{"version":1,"pages":[{"id":"WvWo24gXUlAmIwQMTtvz","title":"What Algenta is","pathname":"/","siteSpaceId":"sitesp_cwO7v","icon":"compass","description":"Algenta is compiler-driven, self-hosted AI infrastructure — 6,000+ deterministic functions on custom Mojo kernels, reached over the HTTP API, the Python and TypeScript SDKs, or MCP. Start here."},{"id":"dC5uSbQw5pwjOTPP9AWU","title":"Install","pathname":"/getting-started/install","siteSpaceId":"sitesp_cwO7v","icon":"download","description":"Install the published Algenta clients or run the self-hosted engine. Pick your path, install in one command where a public package exists, and verify it.","breadcrumbs":[{"label":"Getting Started"}]},{"id":"bx4Gl5he2wFYtNEG4rAq","title":"Quickstart","pathname":"/getting-started/quickstart","siteSpaceId":"sitesp_cwO7v","icon":"bolt","description":"Run Algenta end to end in minutes — install the integrated service, mint an API key, and make your first decision call.","breadcrumbs":[{"label":"Getting Started"}]},{"id":"MCCq9p48DBI7BwBvwixD","title":"Try it without a key","pathname":"/getting-started/playground","siteSpaceId":"sitesp_cwO7v","icon":"flask","description":"Try the Algenta engine with a single POST /v1/playground call — no sign-up and no API key. Run a decision, optimization, simulation, or forecast in one request.","breadcrumbs":[{"label":"Getting Started"}]},{"id":"kcfbhqaXfwEHWksFdtMC","title":"Core concepts","pathname":"/getting-started/concepts","siteSpaceId":"sitesp_cwO7v","icon":"diagram-project","description":"The core Algenta model — one integrated engine, governed data, deterministic decisions, capability routing, and explicit deployment boundaries.","breadcrumbs":[{"label":"Getting Started"}]},{"id":"UvGGKnKB6EqDONoxZRyB","title":"Authentication & API keys","pathname":"/getting-started/authentication","siteSpaceId":"sitesp_cwO7v","icon":"key","description":"Create and use Algenta API keys. Mint a key, send it as a Bearer token, and confirm an authenticated request succeeds.","breadcrumbs":[{"label":"Getting Started"}]},{"id":"yhNYrVe9zjfJTh9Jj7Pl","title":"Verify the bundled engine","pathname":"/engine-runtime/bundled-runtime","siteSpaceId":"sitesp_cwO7v","icon":"microchip","description":"Verify the bundled Algenta compute runtime without installing, starting, or managing a separate engine process.","breadcrumbs":[{"label":"Engine runtime"}]},{"id":"l0L6fUVZOMtx6UivtoX2","title":"Engine health & verification","pathname":"/engine-runtime/health","siteSpaceId":"sitesp_cwO7v","icon":"heart-pulse","description":"Confirm that the API, dependencies, and bundled compute runtime are healthy through supported HTTP probes.","breadcrumbs":[{"label":"Engine runtime"}]},{"id":"7rLNrz1apVPM728hWBgX","title":"Engine catalog","pathname":"/engine-runtime/catalog","siteSpaceId":"sitesp_cwO7v","icon":"layer-group","description":"The Algenta engine catalog — every engine, its slug, API path, tier, and use cases, plus the three endpoints that enumerate them at runtime.","breadcrumbs":[{"label":"Engine runtime"}]},{"id":"AZI3moudrJ1jD8PufsYG","title":"Runtime libraries","pathname":"/engine-runtime/runtime-libraries","siteSpaceId":"sitesp_cwO7v","icon":"cubes","description":"The runtime library surface — list, health-check, and execute individual runtime-backed functions over HTTP, the de CLI, or MCP. Off by default; an operator enables the HTTP routes.","breadcrumbs":[{"label":"Engine runtime"}]},{"id":"kzUdyFwk7apf0Z2fcqj0","title":"Runtime library catalog","pathname":"/engine-runtime/runtime-library-catalog","siteSpaceId":"sitesp_cwO7v","icon":"books","description":"The full catalog of Algenta runtime-library modules — vision, LLM, ML, tensors, statistics, signal, crypto, and more — callable through /v1/libraries.","breadcrumbs":[{"label":"Engine runtime"}]},{"id":"y3y0Sfa7Ud3dt2laDIpo","title":"Kernel inventory","pathname":"/engine-runtime/kernel-inventory","siteSpaceId":"sitesp_cwO7v","icon":"microchip","description":"The public inventory of the engine's CUDA and Metal kernels — what ships, per backend, per model family, and the evidence behind each module.","breadcrumbs":[{"label":"Engine runtime"}]},{"id":"mRZsSheG1CfaMpAdr2hV","title":"Declared capabilities","pathname":"/engine-runtime/capabilities","siteSpaceId":"sitesp_cwO7v","icon":"clipboard-check","description":"Ask the engine what each kernel actually does — deterministic or not, seeded or not, stateful or not, what it costs — and be told plainly when a kernel has not said.","breadcrumbs":[{"label":"Engine runtime"}]},{"id":"yXZ4Y9ieQPEWOGAHXYT7","title":"Batched and chained execution","pathname":"/engine-runtime/coarse-execution","siteSpaceId":"sitesp_cwO7v","icon":"layer-group","description":"Run many kernels — or a whole chain — in one call instead of one call each.","breadcrumbs":[{"label":"Engine runtime"}]},{"id":"5k8ECGUjjJiSyXMbQM64","title":"Reproducibility","pathname":"/engine-runtime/reproducibility","siteSpaceId":"sitesp_cwO7v","icon":"arrows-rotate","description":"Getting the same answer twice — chunk-invariant reductions, one canonical spelling for every float, and a replay manifest that says whether two runs were really the same run.","breadcrumbs":[{"label":"Engine runtime"}]},{"id":"u8Uwo1LQEZTcaYVscvop","title":"Metal LLM decode","pathname":"/gpu-and-llm-inference/qwen3-metal-decode","siteSpaceId":"sitesp_cwO7v","icon":"microchip","description":"Greedy decode on Apple Silicon through the int8 Metal GPU kernel path — call surface, checkpoint preparation, refusals, verification. Currently tested on Qwen3 models.","breadcrumbs":[{"label":"GPU & LLM inference"}]},{"id":"8AeHLgYjQ4bajO34KaJS","title":"Model support matrix","pathname":"/gpu-and-llm-inference/model-support-matrix","siteSpaceId":"sitesp_cwO7v","icon":"table-cells","description":"Model support as declared fact ranges — the capability table, the conform CLI, conformance-proven models, and the typed-refusal boundary.","breadcrumbs":[{"label":"GPU & LLM inference"}]},{"id":"t21m0ouRfFQr7iQS8Ix1","title":"GPU support","pathname":"/gpu-and-llm-inference/gpu-support","siteSpaceId":"sitesp_cwO7v","icon":"microchip","description":"GPU backend support matrix — Metal (default serve path), NVIDIA CUDA (hardware-verified kernel layer and LLM session), ROCm status, and per-lane requirements.","breadcrumbs":[{"label":"GPU & LLM inference"}]},{"id":"Lu0OIKbdmHDTyYuj1p3a","title":"CUDA backend","pathname":"/gpu-and-llm-inference/cuda-backend","siteSpaceId":"sitesp_cwO7v","icon":"microchip","description":"Run the CUDA backend — availability, NVIDIA driver and architecture requirements, the callable entry points, determinism gates, verified model shapes, and the TypeScript SDK path.","breadcrumbs":[{"label":"GPU & LLM inference"}]},{"id":"PqUtaXyCF1vhkWTAGD1J","title":"Data integrity and encoding","pathname":"/compute-and-kernel-primitives/data-integrity-and-encoding","siteSpaceId":"sitesp_cwO7v","icon":"key","description":"Checksums, encoding, keyed hashing, key derivation, a stream cipher and a Merkle tree — compiled primitives, not a general-purpose library wrapper.","breadcrumbs":[{"label":"Compute & kernel primitives"}]},{"id":"3CSDcgcviKG9bfrV9d45","title":"Config and data-parsing kernels","pathname":"/compute-and-kernel-primitives/config-and-data-parsing-kernels","siteSpaceId":"sitesp_cwO7v","icon":"file-code","description":"TOML and YAML config parsing with dot-notation lookups, an RFC 4180 CSV reader, a columnar DataFrame factory, and the zero-Python JSON codec that the engine's own dispatch layer decodes every runtime.","breadcrumbs":[{"label":"Compute & kernel primitives"}]},{"id":"UMHYG3s7MFupdB8n61JU","title":"Compression kernels","pathname":"/compute-and-kernel-primitives/compression-kernels","siteSpaceId":"sitesp_cwO7v","icon":"file-zipper","description":"Delta, Huffman, LZ77 and run-length encoding as compiled kernels — four classic lossless algorithms, each matched to one specific shape of redundancy.","breadcrumbs":[{"label":"Compute & kernel primitives"}]},{"id":"TjGFBTIKRtA2lLGde2Y9","title":"Tensor buffer primitives","pathname":"/compute-and-kernel-primitives/tensorx-kernels","siteSpaceId":"sitesp_cwO7v","icon":"cube","description":"Dtype-tagged byte buffers, elementwise ops, wide-accumulating reductions, N-D shape/broadcasting and low-bit quantization — the flat tensor substrate other kernel families decode through.","breadcrumbs":[{"label":"Compute & kernel primitives"}]},{"id":"UaD70rs7HPgjk7eM6h3z","title":"Normalization kernels","pathname":"/compute-and-kernel-primitives/normalization-kernels","siteSpaceId":"sitesp_cwO7v","icon":"scale-balanced","description":"LayerNorm and RMSNorm forward/backward kernels over flat row-major buffers, plus the shared contract that validates their row shapes and epsilon.","breadcrumbs":[{"label":"Compute & kernel primitives"}]},{"id":"rJH8rrBPxNpkIm09EGzZ","title":"Attention kernels","pathname":"/compute-and-kernel-primitives/attention-kernels","siteSpaceId":"sitesp_cwO7v","icon":"layer-group","description":"Scaled dot-product attention, flash attention, causal/sliding-window masking, RoPE, and analytic backward gradients, in flat row-major layouts with grouped-query support.","breadcrumbs":[{"label":"Compute & kernel primitives"}]},{"id":"xXXg7hKZjTjBUztpuZiJ","title":"MLP block kernels","pathname":"/compute-and-kernel-primitives/mlp-block-kernels","siteSpaceId":"sitesp_cwO7v","icon":"cubes","description":"Gated feed-forward activations (SwiGLU and both GEGLU conventions), token embedding lookup, and the shared feed-forward/vocabulary shape contract, forward and backward, in flat row-major layouts.","breadcrumbs":[{"label":"Compute & kernel primitives"}]},{"id":"EgLQto0w3JCzHBZT21cy","title":"GEMM kernels","pathname":"/compute-and-kernel-primitives/gemm-kernels","siteSpaceId":"sitesp_cwO7v","icon":"table-cells","description":"Cache-blocked SIMD dense GEMM/GEMV in Float64, NF4/int4/int8 quantized matmul with dequantization fused into the reduction loop, and the shared status/dimension/work-budget contract behind both.","breadcrumbs":[{"label":"Compute & kernel primitives"}]},{"id":"kvdghbF0F3m04lxfd7a9","title":"GPU acceleration kernels","pathname":"/compute-and-kernel-primitives/gpu-acceleration-kernels","siteSpaceId":"sitesp_cwO7v","icon":"bolt","description":"Backend selection contract, host device capability probing without compiling a shader, and a threadgroup-tiled GEMM kernel on the Metal GPU.","breadcrumbs":[{"label":"Compute & kernel primitives"}]},{"id":"JUiFCtjqsuMstFCV8bZo","title":"LLM inference kernels","pathname":"/compute-and-kernel-primitives/llm-inference-kernels","siteSpaceId":"sitesp_cwO7v","icon":"bolt","description":"A parallel decode-shaped matrix-vector product, KV-cache growth, greedy argmax, whole-buffer residual add, and the shared status vocabulary that a real autoregressive decode loop composes them into.","breadcrumbs":[{"label":"Compute & kernel primitives"}]},{"id":"5vBW1veTau64pNwEkE35","title":"Attention variant kernels","pathname":"/compute-and-kernel-primitives/transformer-attention-variant-kernels","siteSpaceId":"sitesp_cwO7v","icon":"brain","description":"Scalar reference formulas for scaled dot-product attention, ten named sparse/long-context attention techniques, flash attention's online-softmax building blocks, and five families of positional encodi","breadcrumbs":[{"label":"Compute & kernel primitives"}]},{"id":"qdw2NAS9AXWTevtaxtYc","title":"Transformer blocks and normalization","pathname":"/compute-and-kernel-primitives/transformer-blocks-and-normalization-reference-kernels","siteSpaceId":"sitesp_cwO7v","icon":"diagram-project","description":"Scalar reference formulas for assembling a transformer block (residual, pre/post-norm, causal mask, FFN), a broad normalization catalog beyond LayerNorm/RMSNorm, general non-transformer NN primitives,","breadcrumbs":[{"label":"Compute & kernel primitives"}]},{"id":"YxiSgYKXtdpjynxkrhFj","title":"KV-cache and generation kernels","pathname":"/compute-and-kernel-primitives/kv-cache-and-autoregressive-generation-kernels","siteSpaceId":"sitesp_cwO7v","icon":"repeat","description":"Scalar formulas for KV-cache memory and eviction (plain and paged), sampling and generation-loop decisions, speculative-decoding accept-rate math, and continuous-batching scheduling -- the arithmetic","breadcrumbs":[{"label":"Compute & kernel primitives"}]},{"id":"hK1zMkYSJUCyWph8bE45","title":"Multimodal and inference operations","pathname":"/compute-and-kernel-primitives/multimodal-and-inference-operations-kernels","siteSpaceId":"sitesp_cwO7v","icon":"shield-halved","description":"Scalar analytic formulas for multimodal fusion primitives, inference-serving capacity and latency estimates, per-request cost accounting, and prompt-injection/output-safety scoring.","breadcrumbs":[{"label":"Compute & kernel primitives"}]},{"id":"uHPVJrykjYZRvkGUyBZK","title":"Loss functions and resident-context kernels","pathname":"/compute-and-kernel-primitives/loss-functions-and-resident-context-kernels","siteSpaceId":"sitesp_cwO7v","icon":"gauge-high","description":"Chunked cross-entropy that never materializes the full logit tensor, and the host-side capacity and dispatch-budget arithmetic for a resident GPU decode session, answered with no device required.","breadcrumbs":[{"label":"Compute & kernel primitives"}]},{"id":"0D8LKpd3RqkEt9dGpQim","title":"Neural network primitives and data-pipeline kernels","pathname":"/compute-and-kernel-primitives/neural-network-primitives-and-data-pipeline-kernels","siteSpaceId":"sitesp_cwO7v","icon":"puzzle-piece","description":"Eleven unrelated small libraries closing out the kernel-documentation backlog -- scalar activation functions, cellular automata, a string-similarity column-name ranker, two distinct data-pipeline util","breadcrumbs":[{"label":"Compute & kernel primitives"}]},{"id":"eewPAaocfc9MaRwL7tBY","title":"Core linear algebra","pathname":"/numerical-and-mathematical-computing/core-linear-algebra-kernels","siteSpaceId":"sitesp_cwO7v","icon":"square-root-variable","description":"Dense matrix arithmetic, LU/QR/Cholesky decompositions, 2×2 eigenvalues and power iteration, and elementwise matrix operations, split across two independent in-tree matrix representations.","breadcrumbs":[{"label":"Numerical & mathematical computing"}]},{"id":"qhuh1ksbCW9JRrMcH8JX","title":"Exact numerics and graph kernels","pathname":"/numerical-and-mathematical-computing/exact-numerics-and-graph-kernels","siteSpaceId":"sitesp_cwO7v","icon":"weight-scale","description":"Double-double error-free-transform arithmetic and compensated Householder least squares, plus explicit-graph traversal, shortest paths, MST, and union-find.","breadcrumbs":[{"label":"Numerical & mathematical computing"}]},{"id":"cohUrTUASfFe4k3hkdWN","title":"Sparse structures & tensor primitives","pathname":"/numerical-and-mathematical-computing/sparse-and-tensor-kernels","siteSpaceId":"sitesp_cwO7v","icon":"shapes","description":"Coordinate-format sparse matrices and paired-array sparse vectors, a dense 1D tensor type, and the standalone scalar index/accumulation formulas an N-D tensor implementation composes.","breadcrumbs":[{"label":"Numerical & mathematical computing"}]},{"id":"IFdfE91wyiqlp8U6EKgX","title":"Probability distributions & dependence","pathname":"/numerical-and-mathematical-computing/probability-distributions-and-dependence-kernels","siteSpaceId":"sitesp_cwO7v","icon":"dice","description":"Sampling and MLE/method-of-moments fitting for twenty-six probability distributions, core probability theory (Bayes, combinatorics, entropy), a Gaussian copula with rank-correlation measures, and Chol","breadcrumbs":[{"label":"Numerical & mathematical computing"}]},{"id":"gvuEodE3FuMPrvZR3Ydk","title":"Statistical computing","pathname":"/numerical-and-mathematical-computing/statistical-computing-kernels","siteSpaceId":"sitesp_cwO7v","icon":"dice","description":"Descriptive statistics with a bundled introsort, bootstrap and jackknife resampling for confidence intervals and bias, and Monte Carlo simulation of random walks, diffusions, and point processes.","breadcrumbs":[{"label":"Numerical & mathematical computing"}]},{"id":"p9os0N6OlGOAGeDeahws","title":"Calculus & differential equations","pathname":"/numerical-and-mathematical-computing/calculus-and-differential-equation-kernels","siteSpaceId":"sitesp_cwO7v","icon":"square-root-variable","description":"Interpolation between data points, quadrature over sampled data, initial-value ODE solvers (Euler, RK4), and explicit/Jacobi finite-difference PDE solvers with their CFL stability limits.","breadcrumbs":[{"label":"Numerical & mathematical computing"}]},{"id":"1MogOGBkIGAK0AqDDthk","title":"Polynomial, root-finding & number theory","pathname":"/numerical-and-mathematical-computing/polynomial-root-finding-and-number-theory-kernels","siteSpaceId":"sitesp_cwO7v","icon":"square-root-variable","description":"Dense-coefficient polynomial arithmetic and calculus, five classical root-finding algorithms run against a fixed menu of test functions, and integer number theory -- primality, GCD, modular exponentia","breadcrumbs":[{"label":"Numerical & mathematical computing"}]},{"id":"GBFc3JPdwd3E4tKsuvJB","title":"Combinatorics, convex optimization & scipy-compat","pathname":"/numerical-and-mathematical-computing/combinatorics-convex-optimization-and-scipy-compat-kernels","siteSpaceId":"sitesp_cwO7v","icon":"puzzle-piece","description":"Combinatorial search heuristics, an unconstrained convex-optimization sandbox, a frozen two-variable LP solver, quasi-random sampling, and the scalar formulas behind scipy.optimize and scipy.stats.","breadcrumbs":[{"label":"Numerical & mathematical computing"}]},{"id":"RLOOp3TdjvTfAjV1VaR7","title":"Signal transforms & processing","pathname":"/numerical-and-mathematical-computing/signal-transforms-and-processing-kernels","siteSpaceId":"sitesp_cwO7v","icon":"signal-stream","description":"Radix-2 FFT and Haar wavelet transforms, two independently-implemented frequency-domain feature modules, window/peak/zero-crossing signal utilities, whole-column statistical data transforms, and the f","breadcrumbs":[{"label":"Numerical & mathematical computing"}]},{"id":"z20tR0YgYdSSPw5p7ZME","title":"Dimensionality, information theory & complex systems","pathname":"/numerical-and-mathematical-computing/dimensionality-information-theory-and-complex-systems-kernels","siteSpaceId":"sitesp_cwO7v","icon":"shapes","description":"Six unrelated math primitives grouped by name-similarity rather than shared subject -- business unit-consistency checking, PCA, Shannon information theory, special functions, chaotic dynamical systems","breadcrumbs":[{"label":"Numerical & mathematical computing"}]},{"id":"yDOVshqgx4INI7HQwgDu","title":"Quantum computing & formal logic","pathname":"/numerical-and-mathematical-computing/quantum-computing-and-formal-logic-kernels","siteSpaceId":"sitesp_cwO7v","icon":"atom","description":"Single-qubit state-vector simulation plus closed-form quantum-computing estimators, propositional/boolean logic and truth tables, fuzzy-set membership and defuzzification, gate-level digital circuit s","breadcrumbs":[{"label":"Numerical & mathematical computing"}]},{"id":"f4akaaKaM5ncWeD9vnFl","title":"Geometry and simulation-physics kernels","pathname":"/numerical-and-mathematical-computing/geometry-and-simulation-physics-kernels","siteSpaceId":"sitesp_cwO7v","icon":"shapes","description":"2D Euclidean geometry, rigid-body game-engine kinematics, chemical reaction-rate kinetics, and point-cloud topological data analysis -- four self-contained modules that share no code with each other o","breadcrumbs":[{"label":"Numerical & mathematical computing"}]},{"id":"yQe0T1PHIG0RVIRUpxQo","title":"Time-series model selection","pathname":"/machine-learning/ml-model-selection","siteSpaceId":"sitesp_cwO7v","icon":"clock-rotate-left","description":"Walk-forward, purged, k-fold and temporal splits for leakage-free time-series model selection.","breadcrumbs":[{"label":"Machine learning"}]},{"id":"gShuyMiXIXW4Va2PRajt","title":"Statistical model comparison","pathname":"/machine-learning/ml-model-comparison","siteSpaceId":"sitesp_cwO7v","icon":"scale-balanced","description":"Paired bootstrap CIs, exact McNemar and Diebold-Mariano tests that decide whether a candidate model is genuinely better than the champion.","breadcrumbs":[{"label":"Machine learning"}]},{"id":"5VZvQrXUwRnbtflJQALt","title":"Probability calibration","pathname":"/machine-learning/ml-calibration","siteSpaceId":"sitesp_cwO7v","icon":"sliders","description":"Isotonic (PAV) and z-score affine calibration with reliability diagnostics for turning raw model scores into trustworthy probabilities.","breadcrumbs":[{"label":"Machine learning"}]},{"id":"71DycA3bfUtqPBjAuXlH","title":"Bayesian methods, calibration, and clustering metrics","pathname":"/machine-learning/bayesian-methods-and-calibration-kernels","siteSpaceId":"sitesp_cwO7v","icon":"brain","description":"Conjugate Bayesian updates and Monte Carlo A/B testing, general-purpose variational/Laplace inference primitives, probability calibration, and cluster-quality metrics across four independent kernel mo","breadcrumbs":[{"label":"Machine learning"}]},{"id":"2Oo8FHh7IgAwngVmrSCg","title":"Dense 2D matrices (Matrix2D)","pathname":"/machine-learning/ml-tensor2d","siteSpaceId":"sitesp_cwO7v","icon":"table-list","description":"Dense 2-D matrix type with matmul, matvec and outer products backing the trainable-model kernels.","breadcrumbs":[{"label":"Machine learning"}]},{"id":"NbBHzLwb6uBGx2l1DUAo","title":"LSTM and GRU cells","pathname":"/machine-learning/ml-sequence-cells","siteSpaceId":"sitesp_cwO7v","icon":"code-branch","description":"LSTM and GRU cell forward passes with exact gate math for sequence models.","breadcrumbs":[{"label":"Machine learning"}]},{"id":"kRlyQ7dfPGRMthjeLWEi","title":"LSTM sequence training (BPTT)","pathname":"/machine-learning/ml-sequence-training","siteSpaceId":"sitesp_cwO7v","icon":"gears","description":"Full BPTT training loop for LSTM regressors, gradient-checked against finite differences.","breadcrumbs":[{"label":"Machine learning"}]},{"id":"MpMb4hcpmw2OsQGZU9SM","title":"Bit-exact weight serialization","pathname":"/machine-learning/ml-serialization","siteSpaceId":"sitesp_cwO7v","icon":"database","description":"Bit-exact float64 weight serialization for saving and restoring trained kernels.","breadcrumbs":[{"label":"Machine learning"}]},{"id":"U1BDiMvrqSLqHVvRQBcq","title":"Gradient-boosted regression trees","pathname":"/machine-learning/ml-gradient-boosting","siteSpaceId":"sitesp_cwO7v","icon":"diagram-project","description":"Gradient-boosted regression trees with exact SSE-gain splits, subsampling and shrinkage.","breadcrumbs":[{"label":"Machine learning"}]},{"id":"SJijHDiDefosG8sL6PSN","title":"Gaussian mixture models","pathname":"/machine-learning/ml-mixture-models","siteSpaceId":"sitesp_cwO7v","icon":"chart-scatter","description":"Gaussian mixture models fit by log-space EM with BIC/AIC model selection.","breadcrumbs":[{"label":"Machine learning"}]},{"id":"BdTiMInaENyGxbRpjhd6","title":"Hidden Markov models","pathname":"/machine-learning/ml-hmm","siteSpaceId":"sitesp_cwO7v","icon":"diagram-project","description":"Hidden Markov models — forward/backward likelihood, Viterbi decoding and Baum-Welch training.","breadcrumbs":[{"label":"Machine learning"}]},{"id":"lBiYrLVEa3WUxLkSfdsP","title":"Imbalanced-data resampling","pathname":"/machine-learning/ml-imbalance","siteSpaceId":"sitesp_cwO7v","icon":"scale-balanced","description":"SMOTE, random over/under-sampling and balanced class weights for skewed datasets.","breadcrumbs":[{"label":"Machine learning"}]},{"id":"E5dzXzj1wuMPc6We29gr","title":"Model ensembling","pathname":"/machine-learning/ml-ensembles","siteSpaceId":"sitesp_cwO7v","icon":"layer-group","description":"Stacking and weighted voting ensembles over arbitrary base-model predictions.","breadcrumbs":[{"label":"Machine learning"}]},{"id":"qtbJedm8aUdsHhrHYSem","title":"Conformal prediction intervals","pathname":"/machine-learning/ml-conformal","siteSpaceId":"sitesp_cwO7v","icon":"gauge-high","description":"Split-conformal prediction intervals with finite-sample coverage guarantees.","breadcrumbs":[{"label":"Machine learning"}]},{"id":"bgBMw3Un1sfooJkzUwlm","title":"Model explainability","pathname":"/machine-learning/ml-explainability","siteSpaceId":"sitesp_cwO7v","icon":"magnifying-glass-chart","description":"KernelSHAP and permutation importance for explaining any model's predictions.","breadcrumbs":[{"label":"Machine learning"}]},{"id":"bpIfgmz4hBZC5iTLluGL","title":"Gaussian process regression","pathname":"/machine-learning/ml-gaussian-processes","siteSpaceId":"sitesp_cwO7v","icon":"chart-scatter","description":"Gaussian process regression with RBF and Matern kernels, exact posterior mean and variance.","breadcrumbs":[{"label":"Machine learning"}]},{"id":"CiTB9wWEl2JIUP2K8P2A","title":"Hyperparameter search schedules","pathname":"/machine-learning/ml-automl","siteSpaceId":"sitesp_cwO7v","icon":"robot","description":"Hyperband successive-halving schedules and Latin-hypercube search plans for hyperparameter optimization.","breadcrumbs":[{"label":"Machine learning"}]},{"id":"1EyG2HexSa94Qx5FkdkH","title":"Fit/transform pipelines","pathname":"/machine-learning/ml-pipelines","siteSpaceId":"sitesp_cwO7v","icon":"diagram-project","description":"Composable fit/transform preprocessing pipelines (standardize, min-max, chained) with exact inverses.","breadcrumbs":[{"label":"Machine learning"}]},{"id":"eBnIV5i4Ksd7VWgcGalu","title":"Trainable linear models","pathname":"/machine-learning/ml-linear-models","siteSpaceId":"sitesp_cwO7v","icon":"chart-line","description":"Trainable logistic regression and elastic-net linear models with exact coordinate-descent math.","breadcrumbs":[{"label":"Machine learning"}]},{"id":"91Hp8gL8HFIUztMm1GQF","title":"Symmetric eigendecomposition and thin SVD","pathname":"/machine-learning/ml-eigendecomposition","siteSpaceId":"sitesp_cwO7v","icon":"cubes","description":"Jacobi symmetric eigendecomposition and thin SVD for full-rank spectral computation.","breadcrumbs":[{"label":"Machine learning"}]},{"id":"szE7QqE5ZRsxnqWAdGor","title":"AR/MA estimation (ARIMA building blocks)","pathname":"/machine-learning/ml-arima","siteSpaceId":"sitesp_cwO7v","icon":"chart-line","description":"Durbin-Levinson AR estimation, innovations MA fitting and ARIMA forecasting building blocks.","breadcrumbs":[{"label":"Machine learning"}]},{"id":"5aXZEHcgjOII9IFdhmw3","title":"Feature hashing (the hashing trick)","pathname":"/machine-learning/ml-feature-hashing","siteSpaceId":"sitesp_cwO7v","icon":"bolt","description":"Seeded FNV-1a feature hashing for fixed-width vectorization of unbounded feature names.","breadcrumbs":[{"label":"Machine learning"}]},{"id":"CPl603GLVixdznhSrHEK","title":"Partial least squares regression (PLS1)","pathname":"/machine-learning/ml-pls","siteSpaceId":"sitesp_cwO7v","icon":"chart-line","description":"NIPALS PLS1 regression for collinear and wide (p > n) feature matrices, sklearn-matched.","breadcrumbs":[{"label":"Machine learning"}]},{"id":"h0w5QXZTElC2ryqoaKMF","title":"Trainable SVM classifier (simplified SMO)","pathname":"/machine-learning/ml-svm","siteSpaceId":"sitesp_cwO7v","icon":"scale-balanced","description":"Trainable soft-margin SVM classifier via simplified SMO with linear and RBF kernels.","breadcrumbs":[{"label":"Machine learning"}]},{"id":"Srhg0InFiu24H9SVs5EE","title":"Batched chain greeks","pathname":"/machine-learning/ml-chain-greeks","siteSpaceId":"sitesp_cwO7v","icon":"layer-group","description":"SIMD-batched Black-Scholes price and greeks across a whole option chain in one call, matching the scalar pricing functions to 1e-9.","breadcrumbs":[{"label":"Machine learning"}]},{"id":"J3Q3qliedwDHvqFlwma0","title":"Classical ML algorithms","pathname":"/machine-learning/ml-classical-algorithms","siteSpaceId":"sitesp_cwO7v","icon":"chart-scatter","description":"Distance functions, k-means and k-NN, Gaussian Naive Bayes, preprocessing transforms, evaluation metrics, and a bagged decision-stump random forest across seven independent kernel modules.","breadcrumbs":[{"label":"Machine learning"}]},{"id":"YDmKhPXzexOUJ0sRJLc7","title":"Regression and classification kernels","pathname":"/machine-learning/regression-and-classification-kernels","siteSpaceId":"sitesp_cwO7v","icon":"square-root-variable","description":"Closed-form OLS/ridge regression, a recursive CART regression tree, four SVM kernel functions with Gram-matrix diagnostics, and ten scalar loss functions for classical supervised learning.","breadcrumbs":[{"label":"Machine learning"}]},{"id":"wCYLWXFTxFG7Cp4IxRNC","title":"Autograd and training stability","pathname":"/machine-learning/autograd-and-training-stability-kernels","siteSpaceId":"sitesp_cwO7v","icon":"diagram-project","description":"Reverse-mode autodiff graph bookkeeping, per-operation gradient formulas, ten optimizer update rules, and the loss-scaling/NaN-guard/numerically-safe primitives that keep a training loop from divergin","breadcrumbs":[{"label":"Machine learning"}]},{"id":"aXD8OOP9cezXVeOAOtyz","title":"LoRA adapters, checkpoint loading & distillation","pathname":"/machine-learning/lora-distillation-and-weight-loading-kernels","siteSpaceId":"sitesp_cwO7v","icon":"puzzle-piece","description":"Per-element LoRA/DoRA/prefix/prompt adapter math, sharded and quantized checkpoint-loading arithmetic, and classical distillation-column design equations (not neural-network knowledge distillation).","breadcrumbs":[{"label":"Machine learning"}]},{"id":"qXZjCcjCnPuvqMN3Xw8R","title":"Mixture-of-experts and routing","pathname":"/machine-learning/mixture-of-experts-and-routing-kernels","siteSpaceId":"sitesp_cwO7v","icon":"route","description":"Sparse mixture-of-experts math (gating, capacity, load-balance and z-loss) plus the separate, higher-level layer that routes a whole request to a model, precision, or backend by cost and latency.","breadcrumbs":[{"label":"Machine learning"}]},{"id":"Y8dfN445EeMbOpglbscM","title":"Model lifecycle and interop","pathname":"/machine-learning/model-lifecycle-and-interop-kernels","siteSpaceId":"sitesp_cwO7v","icon":"boxes-stacked","description":"ABI version and layout arithmetic, safetensors/GGUF/ONNX/HuggingFace interop math, data/tensor/pipeline parallelism formulas, and checkpoint/quantization sizing, across four independent Float64 kernel","breadcrumbs":[{"label":"Machine learning"}]},{"id":"waIQZRMEe68M0RXSpFv9","title":"Sequence and state-space models","pathname":"/machine-learning/sequence-and-state-space-model-kernels","siteSpaceId":"sitesp_cwO7v","icon":"repeat","description":"A 2-layer LSTM training loop with full BPTT and Adam, the per-element linear recurrence math behind Mamba/S4/RetNet-style state-space models, and the scalar heuristics a stateful agent runtime scores","breadcrumbs":[{"label":"Machine learning"}]},{"id":"GMFTUfX7ISaiugiAmdRt","title":"RLHF and generative training","pathname":"/machine-learning/rlhf-and-generative-training-kernels","siteSpaceId":"sitesp_cwO7v","icon":"arrows-split-up-and-left","description":"Preference-optimization losses for RLHF/DPO alignment, diffusion noise schedules and denoising steps, InfoNCE-style contrastive losses, one-sample-at-a-time online learning updates, and Pareto/TOPSIS","breadcrumbs":[{"label":"Machine learning"}]},{"id":"2DwEQGb8zaB6QHZOs6Di","title":"Feedback loops and feature engineering","pathname":"/machine-learning/feedback-loops-and-feature-engineering-kernels","siteSpaceId":"sitesp_cwO7v","icon":"arrows-rotate","description":"Scalar scoring functions for the data flywheel (drift, retraining triggers, RLHF reward, calibration) and the scaling/encoding/selection utilities that build model inputs from raw columns.","breadcrumbs":[{"label":"Machine learning"}]},{"id":"QnSaT10CNa2RJzpCba9G","title":"Optimization & decision programs","pathname":"/optimization-and-simulation/optimization","siteSpaceId":"sitesp_cwO7v","icon":"bullseye","description":"Solve linear, integer, assignment and network-flow programs with compiled kernels on your own machine — and be told when no answer exists.","breadcrumbs":[{"label":"Optimization & simulation"}]},{"id":"YSEx985pKOKxJaHuyZzH","title":"Optimization benchmark plane","pathname":"/optimization-and-simulation/optimization-model-formulation","siteSpaceId":"sitesp_cwO7v","icon":"scale-balanced","description":"A flat typed LP/MIP formulation container, its refusal-first validator and x>=0 lowering, a solve bridge that re-audits every answer in the caller's own coordinates, and a judge that scores a solved o","breadcrumbs":[{"label":"Optimization & simulation"}]},{"id":"QOt9YSmYAirDkq9O9BX3","title":"Routing, scheduling & trade-offs","pathname":"/optimization-and-simulation/optimization-heuristics","siteSpaceId":"sitesp_cwO7v","icon":"route","description":"Vehicle routes, schedules, convex quadratic programs, Pareto fronts and decisions under uncertainty — searched on your own machine, reporting the bound you have rather than claiming an optimum.","breadcrumbs":[{"label":"Optimization & simulation"}]},{"id":"fuNKgjZlv0oo89LmvSYr","title":"General-purpose optimization kernels","pathname":"/optimization-and-simulation/optimization-suite-kernels","siteSpaceId":"sitesp_cwO7v","icon":"sliders","description":"Constrained nonlinear optimization by penalty/barrier/Lagrangian methods, four metaheuristic search strategies, scalar root-finding and minimization on polynomials, and ten modern deep-learning optimi","breadcrumbs":[{"label":"Optimization & simulation"}]},{"id":"p0zN2j3UVrUUgjosy9VT","title":"Discrete-event simulation & queueing","pathname":"/optimization-and-simulation/discrete-event-simulation","siteSpaceId":"sitesp_cwO7v","icon":"timer","description":"Discrete-event simulation with a totally-ordered event calendar, resource pools with priority queues, warm-up detection that reports what it discarded","breadcrumbs":[{"label":"Optimization & simulation"}]},{"id":"3VBlJbfrpuOPKXBhwATT","title":"Strategy comparison & simulation","pathname":"/optimization-and-simulation/strategy-simulation","siteSpaceId":"sitesp_cwO7v","icon":"chess-knight","description":"Compare strategies on common random numbers, rank them with the multiplicity corrected, and get \"no significant difference\" as an answer instead of a coin flip.","breadcrumbs":[{"label":"Optimization & simulation"}]},{"id":"X4exagcIHkjl9NE7hsEk","title":"Contest simulation & service timing","pathname":"/optimization-and-simulation/contest-strategy","siteSpaceId":"sitesp_cwO7v","icon":"chess-clock","description":"Simulate N competitors over discrete rounds and compare service-timing options under common random numbers, with simultaneous intervals and an explicit abstain.","breadcrumbs":[{"label":"Optimization & simulation"}]},{"id":"eoeYUvR2Tib8sLqasogh","title":"Reproducible evolutionary search","pathname":"/optimization-and-simulation/reproducible-search","siteSpaceId":"sitesp_cwO7v","icon":"dna","description":"Evolutionary parameter search whose every draw is addressable, so a fitted result replays bit-exactly on any machine and any core count.","breadcrumbs":[{"label":"Optimization & simulation"}]},{"id":"hWQPjATmHLJZPLZic0AF","title":"Reinforcement learning and stochastic simulation kernels","pathname":"/optimization-and-simulation/reinforcement-learning-and-stochastic-simulation-kernels","siteSpaceId":"sitesp_cwO7v","icon":"dice","description":"Tabular Q-learning, SARSA and value iteration over an explicit MDP, a linear-Gaussian Kalman filter, reservoir and weighted stream sampling, and constant-memory Count-Min/HyperLogLog sketches.","breadcrumbs":[{"label":"Optimization & simulation"}]},{"id":"M1lumOJsJs7GYZjWtLcV","title":"Asset degradation state","pathname":"/optimization-and-simulation/degradation-state","siteSpaceId":"sitesp_cwO7v","icon":"battery-three-quarters","description":"Track an asset whose performance decays with usage age — a two-state filter with wear clamping, load correction and a cliff term for the non-linear end of life.","breadcrumbs":[{"label":"Optimization & simulation"}]},{"id":"Q8Mllz18TNE36UDo7GFt","title":"Forecasting and backtesting","pathname":"/time-series-and-statistics/forecasting","siteSpaceId":"sitesp_cwO7v","icon":"arrow-trend-up","description":"Forecast a series, measure the method out of sample, put an honest band around it, and make a hierarchy add up — with a refusal wherever the answer does not exist.","breadcrumbs":[{"label":"Time series & statistics"}]},{"id":"kO7x9Fwz5cQ5CHmFJlxy","title":"ES-RNN residual correction","pathname":"/time-series-and-statistics/ml-esrnn","siteSpaceId":"sitesp_cwO7v","icon":"wave-square","description":"ES-RNN-class residual correction over a frozen exponential-smoothing backbone — backbone selection, a per-series-adaptive window feature, and two correctors that consume it.","breadcrumbs":[{"label":"Time series & statistics"}]},{"id":"k0Eivku4vW8xeRp2ZAjK","title":"Time-series decomposition and queueing kernels","pathname":"/time-series-and-statistics/time-series-decomposition-and-queueing-kernels","siteSpaceId":"sitesp_cwO7v","icon":"hourglass-half","description":"Closed-form M/M/1, M/M/c, and Erlang B/C queueing formulas, plus additive time-series decomposition (trend, seasonal, residual) and a lightweight STL-like iterator -- two unrelated formula libraries s","breadcrumbs":[{"label":"Time series & statistics"}]},{"id":"jjCeZUcZcGkhYXqUbfl3","title":"Forecast scoring & Murphy decomposition","pathname":"/time-series-and-statistics/forecast-scoring","siteSpaceId":"sitesp_cwO7v","icon":"bullseye-arrow","description":"Score probabilistic forecasts honestly — Brier skill, calibration error, and the Murphy decomposition that separates being wrong from being overconfident.","breadcrumbs":[{"label":"Time series & statistics"}]},{"id":"eGKWteQK25Jf8pJ0JYzE","title":"Statistical inference","pathname":"/time-series-and-statistics/statistical-inference","siteSpaceId":"sitesp_cwO7v","icon":"flask","description":"p-values that carry the verdict on their own assumptions, sample sizes verified by re-evaluating power, reproducible resampling, and a critical value that is absent rather than clamped.","breadcrumbs":[{"label":"Time series & statistics"}]},{"id":"dORVeEiBzY2nXhABQBpT","title":"A/B testing and causal inference kernels","pathname":"/time-series-and-statistics/ab-testing-and-causal-inference-kernels","siteSpaceId":"sitesp_cwO7v","icon":"list-check","description":"Two-arm experiment sizing and bandit selection, a compact single-formula causal-effects toolkit, classical decision theory under uncertainty, an ML-to-decision fusion layer, and five nonparametric hyp","breadcrumbs":[{"label":"Time series & statistics"}]},{"id":"SLP4BvzWIEnhUvzdeXe4","title":"Causal inference","pathname":"/time-series-and-statistics/causal-inference","siteSpaceId":"sitesp_cwO7v","icon":"sitemap","description":"Identification first — d-separation and the backdoor criterion on a declared graph — then IPW, matching, doubly-robust and panel estimation, and sensitivity analysis, refusing when not identified.","breadcrumbs":[{"label":"Time series & statistics"}]},{"id":"lVmKrCdps4KkbTTAVuf9","title":"Survival analysis","pathname":"/time-series-and-statistics/survival-analysis","siteSpaceId":"sitesp_cwO7v","icon":"heart-pulse","description":"Time-to-event estimation that respects censoring — Kaplan-Meier with confidence bands, Nelson-Aalen, a kernel-smoothed hazard rate, Cox regression with Efron ties","breadcrumbs":[{"label":"Time series & statistics"}]},{"id":"g02WT3kGK0Il08TMvq7U","title":"Evaluation, survival, and psychometric kernels","pathname":"/time-series-and-statistics/ml-evaluation-survival-and-psychometric-kernels","siteSpaceId":"sitesp_cwO7v","icon":"ruler-combined","description":"Frozen hypothesis-test, survival-curve, operational-metrics, psychometric, and interval-arithmetic primitives that other kernel families call rather than reimplement.","breadcrumbs":[{"label":"Time series & statistics"}]},{"id":"qKdAwjFXLLQuG1xJFs9c","title":"Evaluation & calibration gates","pathname":"/time-series-and-statistics/evaluation-calibration","siteSpaceId":"sitesp_cwO7v","icon":"gauge-simple-high","description":"Brier with Murphy's decomposition, log-loss that says when it was clamped, calibration error that never travels without its binning, reproducible bootstrap intervals","breadcrumbs":[{"label":"Time series & statistics"}]},{"id":"rL1u1SbDIPVohzTxrIDf","title":"Program execution & scoring","pathname":"/time-series-and-statistics/program-execution-scoring","siteSpaceId":"sitesp_cwO7v","icon":"calculator","description":"Executes a resolved FinQA-style arithmetic program — ten operations, steps that reference earlier results by index — to the official evaluator's own rounding spec, bit-for-bit.","breadcrumbs":[{"label":"Time series & statistics"}]},{"id":"uAYcwCI31FQ7zHUBTXFX","title":"Sparse systems & graph analytics","pathname":"/data-and-retrieval/sparse-and-graph","siteSpaceId":"sitesp_cwO7v","icon":"diagram-project","description":"Solve sparse linear systems and analyse graphs with compiled kernels on your own machine.","breadcrumbs":[{"label":"Data & retrieval"}]},{"id":"sZRTrPFZbOkbKcnhU8LN","title":"Retrieval & ranking","pathname":"/data-and-retrieval/retrieval","siteSpaceId":"sitesp_cwO7v","icon":"magnifying-glass","description":"Search, fuse, rerank and explain results with compiled kernels on your own machine — and find out when a backend is missing instead of getting an empty list.","breadcrumbs":[{"label":"Data & retrieval"}]},{"id":"XtDmib1UEomnGUMbUWxn","title":"Retrieval fusion and signal kernels","pathname":"/data-and-retrieval/retrieval-fusion-and-signal-kernels","siteSpaceId":"sitesp_cwO7v","icon":"filter","description":"Reciprocal-rank fusion, hybrid blending, MMR reranking and rank-quality ratios over N ranked lists, plus the moving-average, EMA, IIR and spectral-adjacent filters that operate on one numeric series.","breadcrumbs":[{"label":"Data & retrieval"}]},{"id":"dt9v46YR5hoMQ0QKohEy","title":"Retrieval, recommendation and ranking kernels","pathname":"/data-and-retrieval/retrieval-recommendation-and-ranking-kernels","siteSpaceId":"sitesp_cwO7v","icon":"compass","description":"RAG-specific fusion and quality scores, user-user collaborative filtering, ground-truth ranking metrics, vector-embedding arithmetic, and credit-style scoring formulas — five independent scalar librar","breadcrumbs":[{"label":"Data & retrieval"}]},{"id":"wHVkEt3Utg6dWYycuQJj","title":"Data quality & feature hygiene","pathname":"/data-and-retrieval/data-quality","siteSpaceId":"sitesp_cwO7v","icon":"clipboard-check","description":"Expectations that name the failing row and column, a profile that keeps null, NaN and zero apart, duplicates under a key and an equality you declare","breadcrumbs":[{"label":"Data & retrieval"}]},{"id":"AiJdiSQGIhpwDuDGxQYP","title":"Geospatial indexing & joins","pathname":"/data-and-retrieval/geospatial","siteSpaceId":"sitesp_cwO7v","icon":"map-location-dot","description":"A spatial index with a measured selectivity, a three-valued point-in-polygon with no epsilon, coordinate frames that refuse a swapped latitude, and joins","breadcrumbs":[{"label":"Data & retrieval"}]},{"id":"IvQ87HN8pSeVdLiTigEw","title":"Streaming sketches","pathname":"/data-and-retrieval/streaming-sketches","siteSpaceId":"sitesp_cwO7v","icon":"chart-mixed","description":"Distinct counts, frequencies and quantiles from summaries that fit in constant memory, merge across workers, survive a restart, and report absence rather than zero when there is no answer.","breadcrumbs":[{"label":"Data & retrieval"}]},{"id":"uIvCci00ydcP9boPEtVM","title":"Windowed aggregates","pathname":"/data-and-retrieval/streaming-windows","siteSpaceId":"sitesp_cwO7v","icon":"chart-line-up","description":"Rolling minimum, maximum, range, mean and standard deviation over a sliding window, plus exponentially weighted mean and variance — with the real cost of each one charged before it runs.","breadcrumbs":[{"label":"Data & retrieval"}]},{"id":"nPEDnd1jedjiWePuSrIL","title":"Columnar queries","pathname":"/data-and-retrieval/columnar-queries","siteSpaceId":"sitesp_cwO7v","icon":"table-columns","description":"Filter, group, sort and rank tabular data with compiled kernels on your own machine.","breadcrumbs":[{"label":"Data & retrieval"}]},{"id":"eyemOTyXMGTFPWVoMLZ1","title":"Spreadsheet formula plane","pathname":"/data-and-retrieval/sheet-formula-plane","siteSpaceId":"sitesp_cwO7v","icon":"table-cells","description":"Recalculate real spreadsheet formulas — a tokenizer, dependency graph, cycle detection and 48 census-ranked functions, run as compiled kernels inside the runtime that ships with the SDK.","breadcrumbs":[{"label":"Data & retrieval"}]},{"id":"oGcLNzK91OV69OVeCY3n","title":"Text processing kernels","pathname":"/data-and-retrieval/text-processing-kernels","siteSpaceId":"sitesp_cwO7v","icon":"spell-check","description":"Fuzzy string distance, a hand-rolled regex engine, generic string utilities, word/sentence tokenization and a lexical column-name scorer, all zero-Python-import Mojo.","breadcrumbs":[{"label":"Data & retrieval"}]},{"id":"dVy2oZGvmEeq9OBCO2rY","title":"Text processing and tokenization kernels","pathname":"/data-and-retrieval/text-processing-and-tokenization-kernels","siteSpaceId":"sitesp_cwO7v","icon":"language","description":"Corpus-linguistics readability and diversity statistics, a small from-scratch IR toolkit (TF-IDF, cosine, Jaccard), lexicon-based sentiment scoring, and the arithmetic primitives behind BPE/SentencePi","breadcrumbs":[{"label":"Data & retrieval"}]},{"id":"uV7iO3lOlcm7eRydbH3z","title":"Text similarity and pattern matching kernels","pathname":"/data-and-retrieval/text-similarity-and-pattern-matching-kernels","siteSpaceId":"sitesp_cwO7v","icon":"magnifying-glass","description":"A second, broader string-distance library, a minimal backtracking regex engine distinct from the corpus's other one, and vector-similarity formulas that operate on pre-reduced scalars rather than raw","breadcrumbs":[{"label":"Data & retrieval"}]},{"id":"qznyxRntL9HnwVOdx7xU","title":"Vector ranking kernels","pathname":"/data-and-retrieval/vector-ranking-kernels","siteSpaceId":"sitesp_cwO7v","icon":"ranking-star","description":"SIMD Float32 dot-product and cosine-similarity primitives, a resident candidate-vector table, and the JSON-in/JSON-out column-name and repository-file rankers built on top of them.","breadcrumbs":[{"label":"Data & retrieval"}]},{"id":"ji2wrdy8S5Z2sihFhgEo","title":"Graph-based reasoning","pathname":"/data-and-retrieval/graph-based-reasoning-kernels","siteSpaceId":"sitesp_cwO7v","icon":"circle-nodes","description":"Scalar formulas for compiling an execution graph, scoring decisions walked over a graph under real constraints, and blending ML, simulation and symbolic rules into one reasoning step.","breadcrumbs":[{"label":"Data & retrieval"}]},{"id":"5ZpEMy6QVOroti7q2bYP","title":"Data structures and algorithms kernels","pathname":"/systems-and-infrastructure-kernels/data-structures-and-algorithms-kernels","siteSpaceId":"sitesp_cwO7v","icon":"list-ol","description":"Binary heaps, a priority queue and circular buffer, a byte-level trie, six sorting algorithms, and sorted-list set operations -- pure-Mojo replacements for heapq, queue, pygtrie, sort, and set.","breadcrumbs":[{"label":"Systems & infrastructure kernels"}]},{"id":"2LywSwGtfxTCcDsqtIc3","title":"Bit-level and hashing kernels","pathname":"/systems-and-infrastructure-kernels/bit-level-and-hashing-kernels","siteSpaceId":"sitesp_cwO7v","icon":"fingerprint","description":"Bit-twiddling primitives, non-cryptographic checksums and check digits, a counting Bloom filter, counter-based Philox4x64-10 random draws, and perceptual image hashing for near-duplicate detection.","breadcrumbs":[{"label":"Systems & infrastructure kernels"}]},{"id":"iTqBmiaaOS7wypJHBBjs","title":"Caching and persistence kernels","pathname":"/systems-and-infrastructure-kernels/caching-and-persistence-kernels","siteSpaceId":"sitesp_cwO7v","icon":"database","description":"A TTL cache with hit-rate tracking, an in-memory table/query engine, a key-value store with snapshot/restore, and closed-form memory-sizing arithmetic for pools, allocators, and transformer training.","breadcrumbs":[{"label":"Systems & infrastructure kernels"}]},{"id":"jB3MIezVqDeoOb5RK1T4","title":"Encoding and cryptography kernels","pathname":"/systems-and-infrastructure-kernels/encoding-and-cryptography-kernels","siteSpaceId":"sitesp_cwO7v","icon":"lock","description":"Categorical data encoding, closed-form sizing arithmetic for block-cipher modes, classical-cryptanalysis statistics, and closed-form blockchain/consensus metrics — none of which perform an actual cryp","breadcrumbs":[{"label":"Systems & infrastructure kernels"}]},{"id":"KG1ObTADI2OM7Hc6q3tF","title":"Networking and social graph kernels","pathname":"/systems-and-infrastructure-kernels/networking-and-social-graph-kernels","siteSpaceId":"sitesp_cwO7v","icon":"share-nodes","description":"HTTP/URL protocol primitives, undirected and directed graph-metric arithmetic over adjacency lists, and the scalar cost formulas behind multi-GPU collective communication.","breadcrumbs":[{"label":"Systems & infrastructure kernels"}]},{"id":"jMWlvtBwV3oFAmAu6KkO","title":"Reliability and observability kernels","pathname":"/systems-and-infrastructure-kernels/reliability-and-observability-kernels","siteSpaceId":"sitesp_cwO7v","icon":"shield-halved","description":"Distributed-training checkpoint/recovery math, LLM-serving health metrics, a structured JSON logger, classical MTBF/Weibull reliability engineering, and perturbation-based sensitivity analysis -- five","breadcrumbs":[{"label":"Systems & infrastructure kernels"}]},{"id":"k5xBo8qiaUIoYygGRiTy","title":"Scheduling and rate-limiting kernels","pathname":"/systems-and-infrastructure-kernels/scheduling-and-rate-limiting-kernels","siteSpaceId":"sitesp_cwO7v","icon":"gauge-high","description":"OS-course CPU scheduling algorithms and their metrics, stateless queue/backpressure/rate-limit scoring formulas, three independent token-bucket-family rate limiters, LLM-inference latency micro-optimi","breadcrumbs":[{"label":"Systems & infrastructure kernels"}]},{"id":"BIYjAgbpz4S6KFPlcocU","title":"Governance and workflow kernels","pathname":"/systems-and-infrastructure-kernels/governance-and-workflow-kernels","siteSpaceId":"sitesp_cwO7v","icon":"shield-halved","description":"Petri net formalism primitives, real differential-privacy mechanisms, model/audit governance scoring, ESG carbon-accounting arithmetic, and the scalar primitives behind a DAG workflow orchestrator --","breadcrumbs":[{"label":"Systems & infrastructure kernels"}]},{"id":"nUQ1feAphDCIsiEMnAEA","title":"Utility and control primitive kernels","pathname":"/systems-and-infrastructure-kernels/utility-and-control-primitive-kernels","siteSpaceId":"sitesp_cwO7v","icon":"toolbox","description":"Package-dependency graph resolution, portfolio/Kelly-criterion allocation, statistical anomaly detection, backtracking constraint satisfaction, PID/control-theory step-response math, and numerical (no","breadcrumbs":[{"label":"Systems & infrastructure kernels"}]},{"id":"1ZEVc9plnkWscmdbKQJX","title":"System I/O and serialization kernels","pathname":"/systems-and-infrastructure-kernels/system-io-and-serialization-kernels","siteSpaceId":"sitesp_cwO7v","icon":"toolbox","description":"Environment-variable access, pure-string path and URL parsing, weighted/stratified random sampling, MessagePack binary serialization, and the histogram/dedup pair a richer, mostly-undispatchable conta","breadcrumbs":[{"label":"Systems & infrastructure kernels"}]},{"id":"RO3E0CFD7o63AUcwaAln","title":"Streaming, structured output, and time kernels","pathname":"/systems-and-infrastructure-kernels/streaming-structured-output-and-time-kernels","siteSpaceId":"sitesp_cwO7v","icon":"toolbox","description":"Nine unrelated infrastructure primitives grouped only by being what is left over -- windowed stream aggregates, LLM structured-output scoring, time-zone/date/unit conversion, glob and template string","breadcrumbs":[{"label":"Systems & infrastructure kernels"}]},{"id":"D8ybd82Ajo0GWkxUe7ny","title":"Image feature and quality kernels","pathname":"/vision-and-media-processing/image-feature-and-quality-kernels","siteSpaceId":"sitesp_cwO7v","icon":"image","description":"Sobel edge/saliency features and color histograms, deterministic blur/exposure/entropy quality scores, pixel-level similarity metrics (MSE, PSNR, SSIM), and Otsu-based foreground region boxes over fla","breadcrumbs":[{"label":"Vision & media processing"}]},{"id":"Zw68pHHATFZmji4VDmjx","title":"Image transform and morphology kernels","pathname":"/vision-and-media-processing/image-transform-and-morphology-kernels","siteSpaceId":"sitesp_cwO7v","icon":"image","description":"Binary morphology and connected components, grayscale convolution/filtering, box and bilinear downscaling with integral images, and deterministic RGB palette extraction, all on flat or nested Float64","breadcrumbs":[{"label":"Vision & media processing"}]},{"id":"QuUXE8ANG2eeDctRMM7Z","title":"Color science kernels","pathname":"/vision-and-media-processing/color-science-kernels","siteSpaceId":"sitesp_cwO7v","icon":"palette","description":"RGB/HSL/HSV color-space conversion, CIE colorimetry (delta-E, Lab-to-LCH, CCT, CRI, whiteness), and WCAG text-contrast accessibility checking, layered as three modules of increasing specificity.","breadcrumbs":[{"label":"Vision & media processing"}]},{"id":"05sYOS5ZjerTj6rGpdOm","title":"Document vision and layout kernels","pathname":"/vision-and-media-processing/document-vision-kernels","siteSpaceId":"sitesp_cwO7v","description":"Pixel-level document image processing -- Otsu and adaptive binarization, skew and scan-quality scoring, and projection-based segmentation into text lines, columns, density grids, and table-likeness sc","breadcrumbs":[{"label":"Vision & media processing"}]},{"id":"0TJLdq7QWp0ypZbePmIu","title":"Video quality and accessibility kernels","pathname":"/vision-and-media-processing/video-quality-and-accessibility-kernels","siteSpaceId":"sitesp_cwO7v","icon":"closed-captioning","description":"Per-pixel worst-case WCAG contrast for caption overlays, dominant/mean color analysis, a composite no-reference frame-quality score, and clip-to-clip duplicate detection, all over flat row-major video","breadcrumbs":[{"label":"Vision & media processing"}]},{"id":"VqXlluxm405Gi6X7Gg8K","title":"Video content and scene kernels","pathname":"/vision-and-media-processing/video-content-and-scene-kernels","siteSpaceId":"sitesp_cwO7v","icon":"clapperboard","description":"Frame-level video primitives -- scene-change scoring, motion-region extraction, storyboard/keyframe selection, and the routing gate that decides how much of a clip is worth a model call, mirroring the","breadcrumbs":[{"label":"Vision & media processing"}]},{"id":"OGFx5TrXWH6z6CsZqi2d","title":"Ranging and remote-sensing kernels","pathname":"/vision-and-media-processing/ranging-and-remote-sensing-kernels","siteSpaceId":"sitesp_cwO7v","icon":"satellite-dish","description":"LiDAR point-cloud geometry and vegetation-return metrics, LiDAR-derived terrain/canopy mapping and survey QA, and the core radar range/Doppler/link-budget equations.","breadcrumbs":[{"label":"Vision & media processing"}]},{"id":"eisqI0PbO0CWUJjKlF3J","title":"Solid-Earth geophysics","pathname":"/earth-and-planetary-sciences/solid-earth-geophysics-kernels","siteSpaceId":"sitesp_cwO7v","icon":"mountain","description":"WGS84 geodesy and coordinate conversion, gravity/magnetic/seismic-velocity survey formulas, earthquake magnitude and travel-time relations, and simplified volcanic-eruption dynamics.","breadcrumbs":[{"label":"Earth & planetary sciences"}]},{"id":"Y6QmfBYhMLGHlaEBiHab","title":"Surface processes & geomorphology","pathname":"/earth-and-planetary-sciences/surface-processes-and-geomorphology-kernels","siteSpaceId":"sitesp_cwO7v","icon":"mountain","description":"Landform evolution and fluvial morphometrics, soil/rock mechanics for construction, soil formation and classification science, and sediment transport/grain-size statistics, across four independent sca","breadcrumbs":[{"label":"Earth & planetary sciences"}]},{"id":"NgP28Epq9yyRwTAi7XRx","title":"Hydrology & water systems","pathname":"/earth-and-planetary-sciences/hydrology-and-water-systems-kernels","siteSpaceId":"sitesp_cwO7v","icon":"droplet","description":"Surface hydraulics, groundwater flow, lake physics and chemistry, ocean dynamics, storm-water meteorology and the vegetation-water interface across seven independent kernel modules.","breadcrumbs":[{"label":"Earth & planetary sciences"}]},{"id":"Yv0yK8dRJYrZlkiMTsJs","title":"Historical geology & paleoclimate","pathname":"/earth-and-planetary-sciences/historical-geology-and-paleoclimate-kernels","siteSpaceId":"sitesp_cwO7v","icon":"clock-rotate-left","description":"Radiometric dating, rock-magnetic paleolatitude, sediment-layer age modeling, and ice-sheet mass balance across four independent Earth-history kernel modules.","breadcrumbs":[{"label":"Earth & planetary sciences"}]},{"id":"ZPOcJECACV4ekRNCkGKu","title":"Speleology, mineralogy & geochemistry","pathname":"/earth-and-planetary-sciences/speleology-mineralogy-and-geochemistry-kernels","siteSpaceId":"sitesp_cwO7v","icon":"gem","description":"Cave hydrology, speleothem dating, and cave-radio physics; mineral identification and crystal chemistry; X-ray diffraction and unit-cell geometry; and the trace-element, isotope, and elemental-cycling","breadcrumbs":[{"label":"Earth & planetary sciences"}]},{"id":"EZ5m3fi3OWrhg5gWXvWK","title":"Cartography, planetary & cosmic sciences","pathname":"/earth-and-planetary-sciences/cartography-and-planetary-cosmic-sciences-kernels","siteSpaceId":"sitesp_cwO7v","icon":"satellite","description":"Map-scale and projection arithmetic for cartography, Julian-date and orbital-mechanics formulas for astronomy, Hubble-law cosmology and blackbody radiation laws, and habitability/biosignature scoring","breadcrumbs":[{"label":"Earth & planetary sciences"}]},{"id":"e3ezsHTMSmq8resHMGXG","title":"Atmospheric chemistry, climate & meteorology","pathname":"/earth-and-planetary-sciences/atmospheric-chemistry-and-climate-kernels","siteSpaceId":"sitesp_cwO7v","icon":"smog","description":"Gas-phase photochemistry and pollutant-lifetime formulas, plus two independently-implemented single-point weather-index toolkits that compute instantaneous psychrometric conditions rather than long-te","breadcrumbs":[{"label":"Earth & planetary sciences"}]},{"id":"OoKgzhGs6cGufwRENWt7","title":"Earth, electrical, and neurophysics kernels","pathname":"/earth-and-planetary-sciences/earth-electrical-and-neurophysics-kernels","siteSpaceId":"sitesp_cwO7v","icon":"globe","description":"Spherical-Earth navigation math, textbook DC/AC circuit-theory formulas, and the biophysics of neuron membranes and action potentials, across three unrelated small Float64 kernel modules.","breadcrumbs":[{"label":"Earth & planetary sciences"}]},{"id":"MZUXz9opONosJhleWr5T","title":"Acoustics, optics & radiative materials","pathname":"/physical-and-materials-science/acoustics-optics-and-radiative-materials-kernels","siteSpaceId":"sitesp_cwO7v","description":"Five independent physics-formula modules for sound and light — decibel levels and room acoustics, medical/industrial ultrasound imaging physics, blackbody and photon radiometry, fiber/optical photonic","breadcrumbs":[{"label":"Physical & materials science"}]},{"id":"QcETdqXOuxVqCfnWdoFe","title":"Structural & construction materials","pathname":"/physical-and-materials-science/structural-and-construction-materials-kernels","siteSpaceId":"sitesp_cwO7v","icon":"trowel-bricks","description":"Adhesive bond strength and cure kinetics, fiber-composite mechanics, concrete mix design and curing, ceramic fracture statistics, polymer chain/molecular-weight science, and rubber/elastomer testing,","breadcrumbs":[{"label":"Physical & materials science"}]},{"id":"deVejHIjzxDOgU0g65Rm","title":"Continuum and structural mechanics kernels","pathname":"/physical-and-materials-science/continuum-and-structural-mechanics-kernels","siteSpaceId":"sitesp_cwO7v","icon":"compass-drafting","description":"Classical point mechanics, beam/column structural analysis, Newtonian pipe-and-body fluid dynamics, non-Newtonian and viscoelastic rheology, and soil engineering split across its geotechnical and agro","breadcrumbs":[{"label":"Physical & materials science"}]},{"id":"KOKHsEhVCKJn6uzZm3l1","title":"Corrosion & electrochemical materials","pathname":"/physical-and-materials-science/corrosion-and-electrochemical-materials-kernels","siteSpaceId":"sitesp_cwO7v","icon":"bolt","description":"Corrosion rate and cathodic protection, electrochemical cell and battery equations, classical electrostatics and magnetism, metallurgical strengthening/hardenability, and semiconductor device physics","breadcrumbs":[{"label":"Physical & materials science"}]},{"id":"I0IbqRHkJ3xuK0Vh8yxs","title":"Thermal, cryogenic & energy materials","pathname":"/physical-and-materials-science/thermal-and-energy-materials-kernels","siteSpaceId":"sitesp_cwO7v","icon":"temperature-half","description":"Cryogenic and permafrost thermal physics, the four classical thermodynamic relations, battery electrochemistry, and solar/wind/photovoltaic energy-yield engineering across seven independent kernel mod","breadcrumbs":[{"label":"Physical & materials science"}]},{"id":"iCqR4zkEWjMFlSZ77XwP","title":"Nanoscale, plasma & surface-contact engineering","pathname":"/physical-and-materials-science/advanced-nano-and-plasma-materials-kernels","siteSpaceId":"sitesp_cwO7v","icon":"atom","description":"Nanoscale surface/transport physics, plasma and fusion parameters, dry and lubricated contact mechanics with wear-corrosion coupling, and antenna/RF link-budget equations across five independent kerne","breadcrumbs":[{"label":"Physical & materials science"}]},{"id":"kgbczJbJAOLJEcTKkLo9","title":"Process & organic-materials science","pathname":"/physical-and-materials-science/process-and-organic-materials-science-kernels","siteSpaceId":"sitesp_cwO7v","icon":"industry","description":"Small scalar-formula libraries for eight process and organic-materials domains — brewing, bioprocess fermentation, food science, fragrance chemistry, pulp and paper, timber, textiles, and combustion.","breadcrumbs":[{"label":"Physical & materials science"}]},{"id":"MgIHyUdvpYzKsrpkunbk","title":"Materials analysis and spectroscopy kernels","pathname":"/physical-and-materials-science/materials-analysis-and-spectroscopy-kernels","siteSpaceId":"sitesp_cwO7v","icon":"flask-vial","description":"Generic mechanics-of-materials formulas, chromatographic separation science, optical/analytical spectroscopy, treatment-plant water chemistry, and structural timber engineering across five independent","breadcrumbs":[{"label":"Physical & materials science"}]},{"id":"OODNRBzPrUWQzxGQtHL5","title":"Clinical specialty kernels","pathname":"/medical-and-health-sciences/clinical-specialty-kernels","siteSpaceId":"sitesp_cwO7v","icon":"stethoscope","description":"Textbook clinical formulas for seven specialties -- cardiac hemodynamics, dental/oral-health indices, hormone physiology, kidney function, eye optics, orthodontic biomechanics, and lung function -- as","breadcrumbs":[{"label":"Medical & health sciences"}]},{"id":"vCoBBCVl9HKqX0xEt6u0","title":"Immunology, hematology, and virology kernels","pathname":"/medical-and-health-sciences/immunology-hematology-and-virology-kernels","siteSpaceId":"sitesp_cwO7v","icon":"microscope","description":"Antibody/immune-response, blood-cell-index, viral-kinetics, cell-biology, and flow-cytometry formulas across five independent clinical and lab-science modules.","breadcrumbs":[{"label":"Medical & health sciences"}]},{"id":"pv3FABUhl8m0V6JRJPAX","title":"Pharmacology and toxicology kernels","pathname":"/medical-and-health-sciences/pharmacology-and-toxicology-kernels","siteSpaceId":"sitesp_cwO7v","icon":"vial","description":"One-compartment drug pharmacokinetics and dose-response pharmacology, CNS-specific receptor math, the parallel toxicokinetics/toxicology pair for toxic and environmental substances, and radiation dosi","breadcrumbs":[{"label":"Medical & health sciences"}]},{"id":"iKOgf39hZWlDat9dRoNJ","title":"Epidemiology and disease dynamics kernels","pathname":"/medical-and-health-sciences/epidemiology-and-disease-dynamics-kernels","siteSpaceId":"sitesp_cwO7v","icon":"virus","description":"Compartmental disease-spread models and outbreak metrics, the risk-factor and diagnostic-test statistics used to argue a cause, and molecular assay quantification for gene regulation that never touche","breadcrumbs":[{"label":"Medical & health sciences"}]},{"id":"fUVhVUQMKRhMSnYuOVDU","title":"Biomedical, biomechanics, and psychoacoustics kernels","pathname":"/medical-and-health-sciences/biomedical-engineering-kernels","siteSpaceId":"sitesp_cwO7v","icon":"stethoscope","description":"Bedside cardiovascular/renal/respiratory/pharmacokinetic formulas, gait and joint-mechanics equations, and human sound-perception models — three unrelated small physics/physiology libraries, not one \"","breadcrumbs":[{"label":"Medical & health sciences"}]},{"id":"xXF8H3Qz4mwhNcTEyjz4","title":"Nutrition and veterinary science kernels","pathname":"/medical-and-health-sciences/nutrition-and-veterinary-science-kernels","siteSpaceId":"sitesp_cwO7v","icon":"weight-scale","description":"Human energy-balance and body-composition formulas (BMI, Mifflin-St Jeor BMR, TDEE, macros) alongside weight-based veterinary dosing, fluid therapy, and per-species reference constants for dogs, cats,","breadcrumbs":[{"label":"Medical & health sciences"}]},{"id":"8rEbEZJ86VC0CRsBUMPv","title":"Agricultural production kernels","pathname":"/biology-and-agricultural-sciences/agricultural-production-kernels","siteSpaceId":"sitesp_cwO7v","icon":"seedling","description":"Five flat-scalar modules for crop and plant production math -- field agriculture, agronomic crop/soil science, horticultural plant physiology, hydroponic nutrient/climate control, and greenhouse struc","breadcrumbs":[{"label":"Biology & agricultural sciences"}]},{"id":"8JooqDVhoMBwMjXFwQip","title":"Specialty agriculture & aquaculture kernels","pathname":"/biology-and-agricultural-sciences/specialty-agriculture-and-aquaculture-kernels","siteSpaceId":"sitesp_cwO7v","icon":"seedling","description":"Per-hive, per-tank, per-stock, per-hectare, and per-trap arithmetic for six specialty production sciences -- beekeeping, aquatic farming, wild-fisheries stock assessment, wine-grape growing and cellar","breadcrumbs":[{"label":"Biology & agricultural sciences"}]},{"id":"WJTQwZRgUUGQMKMD6CL8","title":"Molecular & cell biology kernels","pathname":"/biology-and-agricultural-sciences/molecular-and-cell-biology-kernels","siteSpaceId":"sitesp_cwO7v","icon":"leaf","description":"Sequence-level bioinformatics, genome-scale statistics, mass-spectrometry protein and metabolite quantification, allele-frequency population genetics, and plant secondary-metabolite assay formulas acr","breadcrumbs":[{"label":"Biology & agricultural sciences"}]},{"id":"u3jCAHJq3Qbm1kIMozta","title":"Ecology & organismal biology kernels","pathname":"/biology-and-agricultural-sciences/ecology-and-organismal-biology-kernels","siteSpaceId":"sitesp_cwO7v","icon":"bug","description":"Population-dynamics and community-diversity formulas for general ecology, plus five taxon-focused libraries -- ecophysiology, entomology, microbiology, mycology, and bryology -- covering organism-envi","breadcrumbs":[{"label":"Biology & agricultural sciences"}]},{"id":"UXuTefA8Qj7cZq8HVXBB","title":"Paleobiology & archaeobotany kernels","pathname":"/biology-and-agricultural-sciences/paleobiology-and-archaeobotany-kernels","siteSpaceId":"sitesp_cwO7v","icon":"tree","description":"Tree-ring dating and climate-signal statistics, forest-mensuration biometrics, quantitative pollen/spore analysis, and fossil preservation and decay-process formulas across four independent historical","breadcrumbs":[{"label":"Biology & agricultural sciences"}]},{"id":"Zwa9hEhd87072i6B7NCE","title":"Aerospace & autonomous systems kernels","pathname":"/engineering-and-industrial-systems/aerospace-and-autonomous-systems-kernels","siteSpaceId":"sitesp_cwO7v","icon":"plane","description":"Orbital and rocket mechanics plus atmospheric aerodynamics (aerospace), aircraft-specific performance (aviation), UAV power/range/imaging (drones), self-driving-car safety physics (autonomous_vehicles","breadcrumbs":[{"label":"Engineering & industrial systems"}]},{"id":"iNsDlBjWxMIcFzYQImIu","title":"Building & facility systems kernels","pathname":"/engineering-and-industrial-systems/building-and-facility-systems-kernels","siteSpaceId":"sitesp_cwO7v","icon":"building","description":"Vertical-transport and building-facility engineering -- elevator dispatch/sizing, fire-safety egress and suppression, HVAC heat load, plumbing hydraulics, refrigeration cycles, wastewater treatment de","breadcrumbs":[{"label":"Engineering & industrial systems"}]},{"id":"kpOKg0ik4JUyQIPcA3Ct","title":"Industrial fluid power & energy systems kernels","pathname":"/engineering-and-industrial-systems/industrial-fluid-power-and-energy-systems-kernels","siteSpaceId":"sitesp_cwO7v","icon":"industry","description":"Hydraulic (pressurized-liquid) power transmission, pneumatic (compressed-air) power transmission, and grid-scale AC power engineering — three independent kernel modules, none of which overlap with the","breadcrumbs":[{"label":"Engineering & industrial systems"}]},{"id":"qSLif8qDJ8s6F8lmWQWv","title":"Transportation & hazard-safety kernels","pathname":"/engineering-and-industrial-systems/transportation-and-hazard-safety-kernels","siteSpaceId":"sitesp_cwO7v","icon":"traffic-light","description":"Traffic-signal timing and level-of-service formulas, physical oceanography, mine-planning economics with blasting and ventilation arithmetic, and nuclear/reactor decay physics — four independent engin","breadcrumbs":[{"label":"Engineering & industrial systems"}]},{"id":"docH6b2vhWr98pYd9DBA","title":"Manufacturing & craft engineering kernels","pathname":"/engineering-and-industrial-systems/manufacturing-and-craft-engineering-kernels","siteSpaceId":"sitesp_cwO7v","icon":"screwdriver-wrench","description":"Joint-strength and heat-control physics for arc welding, mechanical-watch gear trains and escapement timing, press production math for printing, corrugated-box and container engineering, and textile-c","breadcrumbs":[{"label":"Engineering & industrial systems"}]},{"id":"Ncrad0s6B2Vd48EzAN91","title":"Accounting, actuarial, insurance & pension kernels","pathname":"/finance-business-and-economics/accounting-and-actuarial-kernels","siteSpaceId":"sitesp_cwO7v","icon":"sack-dollar","description":"Depreciation, financial ratios, and time-value-of-money accounting; actuarial mortality, life-expectancy, and reserve math; insurance premium ratemaking and loss-ratio formulas; and pension funding, v","breadcrumbs":[{"label":"Finance, business & economics"}]},{"id":"4EFHRu7NNdMUpLpHJkZr","title":"Auction & market mechanism kernels","pathname":"/finance-business-and-economics/auction-and-market-mechanism-kernels","siteSpaceId":"sitesp_cwO7v","icon":"gavel","description":"Auction revenue and payment rules (Vickrey/English/Dutch/VCG), closed-form behavioral-economics biases, and voting rules computed directly from raw ballots.","breadcrumbs":[{"label":"Finance, business & economics"}]},{"id":"YwAR0TwyCQLA66RPQoER","title":"Macroeconomic indicators & currency kernels","pathname":"/finance-business-and-economics/macroeconomic-indicators-and-currency-kernels","siteSpaceId":"sitesp_cwO7v","icon":"sack-dollar","description":"Time-value-of-money and loan math, corporate/investment finance ratios, national-economy indicators (GDP, inflation, unemployment), and human population statistics -- four independent formula librarie","breadcrumbs":[{"label":"Finance, business & economics"}]},{"id":"TZ8NQy4MrJazo3NL6r4F","title":"Risk, supply-chain & logistics kernels","pathname":"/finance-business-and-economics/risk-supply-chain-and-logistics-kernels","siteSpaceId":"sitesp_cwO7v","icon":"truck","description":"Portfolio/returns-series risk metrics (VaR, Sharpe, drawdown) alongside five independent operations-research modules for demand forecasting, EOQ/safety-stock, procurement economics, logistics KPIs, an","breadcrumbs":[{"label":"Finance, business & economics"}]},{"id":"sdyY2T9tGb3RsHVLt6oA","title":"Business operations & real-estate kernels","pathname":"/finance-business-and-economics/business-operations-and-real-estate-kernels","siteSpaceId":"sitesp_cwO7v","icon":"briefcase","description":"Queueing/inventory/LP-scoring formulas, earned-value and PERT project math, real-estate underwriting ratios and mortgage amortization, and telecom link-budget/traffic-engineering formulas -- four inde","breadcrumbs":[{"label":"Finance, business & economics"}]},{"id":"wJU7anygdLMZj7a2Kqoi","title":"Collectibles, gaming & sports analytics kernels","pathname":"/finance-business-and-economics/collectibles-gaming-and-sports-analytics-kernels","siteSpaceId":"sitesp_cwO7v","icon":"dice","description":"Museum conservation and collections-management arithmetic, coin valuation and grading, lottery combinatorics and expected-value math, and sports performance metrics, across four independent scalar-for","breadcrumbs":[{"label":"Finance, business & economics"}]},{"id":"mtNBdRSkOjwvyxzK7TAl","title":"Humanities and social science kernels","pathname":"/humanities-and-social-sciences/humanities-and-social-science-kernels","siteSpaceId":"sitesp_cwO7v","icon":"landmark","description":"Radiocarbon/stratigraphic dating and artifact statistics for archaeology, NIOSH/Fitts's-law/Hick's-law human-factors formulas for ergonomics, time-of-death/toxicology/ballistics calculations for foren","breadcrumbs":[{"label":"Humanities & social sciences"}]},{"id":"WxWxFc47iyZNfwK18Gl7","title":"How the engine works","pathname":"/concepts/engine","siteSpaceId":"sitesp_cwO7v","icon":"gears","description":"How Algenta integrates deterministic compiled compute, validates every request, and fails closed without exposing runtime lifecycle to users.","breadcrumbs":[{"label":"Concepts"}]},{"id":"qhUF37lxWlpmrmAe0o2e","title":"The governed decision lifecycle","pathname":"/concepts/decision-service","siteSpaceId":"sitesp_cwO7v","icon":"scale-balanced","description":"How Algenta turns enterprise data and operating models into governed, versioned, reproducible decisions instead of one-off chat answers","breadcrumbs":[{"label":"Concepts"}]},{"id":"Lp5WFnhBPXwlBi5l0A2d","title":"Governed data & query","pathname":"/concepts/governed-data","siteSpaceId":"sitesp_cwO7v","icon":"shield-check","description":"Governed data is data Algenta queries under a typed, schema-validated, deterministic, auditable contract — sources, datasets, connectors, and the resolve to query flow.","breadcrumbs":[{"label":"Concepts"}]},{"id":"bjkEhVUT4VwGkNpsxVeo","title":"The capability plane","pathname":"/concepts/capability-plane","siteSpaceId":"sitesp_cwO7v","icon":"diagram-project","description":"The unified router that turns an objective into the right provider and skill, then executes it — discovery, routing, execution, adapters, and fail-closed execution ownership.","breadcrumbs":[{"label":"Concepts"}]},{"id":"kzBRwbzvypUVd7HsDqzw","title":"Simulation models & distributions","pathname":"/concepts/simulation-models","siteSpaceId":"sitesp_cwO7v","icon":"chart-scatter","description":"The simulation algorithms and input distributions Algenta runs — auto vs expert mode, when to pick each engine, and how a seed makes every run reproducible.","breadcrumbs":[{"label":"Concepts"}]},{"id":"ktXIojcv61EsP0IhlJTb","title":"Decision memory","pathname":"/concepts/decision-memory","siteSpaceId":"sitesp_cwO7v","icon":"clock-rotate-left","description":"Decision memory is the persisted log of decisions, the real-world outcomes recorded against them, and the review loop that lets the engine learn which decision types are over- or under-confident.","breadcrumbs":[{"label":"Concepts"}]},{"id":"GjKO6uzgRb2UiU7OiDZN","title":"Execution ownership & fail-closed","pathname":"/concepts/execution-ownership","siteSpaceId":"sitesp_cwO7v","icon":"shield-halved","description":"Who runs a capability — algenta_managed (the engine executes) vs client_managed (your adapter executes) — and why an unresolved or unsupported owner fails closed.","breadcrumbs":[{"label":"Concepts"}]},{"id":"pJGCOrvzSg34BTTmhYkb","title":"Deployment modes & privacy","pathname":"/concepts/deployment-modes","siteSpaceId":"sitesp_cwO7v","icon":"shield-halved","description":"How ALGENTA_DEPLOYMENT_MODE selects a privacy profile — cloud, telemetry, metering, and egress defaults — and how the engine fails closed on outbound network in air-gapped runs.","breadcrumbs":[{"label":"Concepts"}]},{"id":"bxnYz4ESKz4Thr7Unlxd","title":"Make your first query","pathname":"/guides/first-query","siteSpaceId":"sitesp_cwO7v","icon":"magnifying-glass","description":"Run your first governed query against a registered dataset and read the deterministic result, using HTTP, the de CLI, or the Python SDK.","breadcrumbs":[{"label":"Guides"}]},{"id":"iJ8hP9puYHhT8U9OlqPL","title":"Group data locally with Query.group_by()","pathname":"/guides/local-query-groupby","siteSpaceId":"sitesp_cwO7v","icon":"table-cells","description":"Build a columnar query with algenta.Query and group your own data locally with group_by() — the same composed engine the dataframe group-by benchmark measures.","breadcrumbs":[{"label":"Guides"}]},{"id":"eDDxcUfD56Kbb5TWZoYP","title":"Run a decision","pathname":"/guides/run-a-decision","siteSpaceId":"sitesp_cwO7v","icon":"scale-balanced","description":"Turn a scenario into a recommended action — call simulate or recommend and read the decision envelope (recommended action, expected value, confidence, percentiles).","breadcrumbs":[{"label":"Guides"}]},{"id":"LW5Nv4MbrOOLp8FsJrsH","title":"Drive a governed decision","pathname":"/guides/governed-decisions","siteSpaceId":"sitesp_cwO7v","icon":"diagram-project","description":"Frame a decision case, run the governed analysis, inspect the evidence, then propose, approve, and execute an action plan through the host-independent Decision API","breadcrumbs":[{"label":"Guides"}]},{"id":"dDv6ojr8R8nL5SzjPWT7","title":"Renewal Capacity Allocation pack","pathname":"/guides/renewal-capacity-allocation","siteSpaceId":"sitesp_cwO7v","icon":"chart-line","description":"Allocate limited customer-success hours across upcoming renewals to maximize model-estimated retained ARR, modeling intervention effects explicitly under hard constraints.","breadcrumbs":[{"label":"Guides"}]},{"id":"xo3kyzyPZXFb0ETopz2h","title":"Simulate a scenario","pathname":"/guides/simulate","siteSpaceId":"sitesp_cwO7v","icon":"chart-scatter","description":"Build a full simulation request — auto vs expert mode, choose a simulation_model, declare distributions and params, pin a seed, and read the percentiles back.","breadcrumbs":[{"label":"Guides"}]},{"id":"5o844lphUoPLQBkoWgGq","title":"Kernels on Hugging Face","pathname":"/guides/kernels-on-hugging-face","siteSpaceId":"sitesp_cwO7v","icon":"cubes","description":"Load Algenta compute kernels on Apple silicon straight from the Hugging Face Kernel Hub with the kernels library.","breadcrumbs":[{"label":"Guides"}]},{"id":"CH2L7bZyy11m0bzX2E7J","title":"Run a what-if sweep","pathname":"/guides/what-if","siteSpaceId":"sitesp_cwO7v","icon":"sliders","description":"Sweep one simulation variable across a range to see how outcomes move, and stream a long simulation live over Server-Sent Events.","breadcrumbs":[{"label":"Guides"}]},{"id":"AharbnfV1rQPPkNynKqR","title":"Recommend, score & compare","pathname":"/guides/recommend-and-score","siteSpaceId":"sitesp_cwO7v","icon":"ranking-star","description":"Compare named actions and get a ranked recommendation, score a single scenario with custom weights, run many simulations at once, and compare scenarios side by side.","breadcrumbs":[{"label":"Guides"}]},{"id":"1hWq4uaSOQ9pNlD9Ox0f","title":"Query governed data","pathname":"/guides/governed-query","siteSpaceId":"sitesp_cwO7v","icon":"table-list","description":"Run a deterministic governed query end to end — resolve structured intent into an exact plan, execute it with typed filters, and read plan_hash and validated.","breadcrumbs":[{"label":"Guides"}]},{"id":"tGlnNUppZSD65MAWJvD0","title":"Log decisions & record outcomes","pathname":"/guides/decision-memory","siteSpaceId":"sitesp_cwO7v","icon":"clock-rotate-left","description":"Log a decision through the engine, record its real-world outcome later, and review the closed loop so the engine learns which decision types are over or under confident.","breadcrumbs":[{"label":"Guides"}]},{"id":"JeLFg5v7KvcJzEJh1uMb","title":"Drive decisions through the API","pathname":"/guides/decision-memory-api","siteSpaceId":"sitesp_cwO7v","icon":"wand-magic-sparkles","description":"Drive the full Algenta decisions API — plan, log, execute, and delete decisions — and call the one-line product endpoints for decision, agent, optimize, retrieve, and forecast.","breadcrumbs":[{"label":"Guides"}]},{"id":"Dl1erNpIZw5dJvYsl5y2","title":"Configure the execution policy","pathname":"/guides/execution-policy","siteSpaceId":"sitesp_cwO7v","icon":"shield-halved","description":"Read and update the Algenta execution policy — the confidence, risk, calibration, and idempotency gates that guard autonomous decision execution — and inspect calibration.","breadcrumbs":[{"label":"Guides"}]},{"id":"oWVTrgfD6zfzaggU9IRO","title":"Connect a data source","pathname":"/guides/connect-data","siteSpaceId":"sitesp_cwO7v","icon":"plug","description":"Register an external source as a saved connector, test it, browse its schema, and onboard a dataset so the engine can query it under a typed contract.","breadcrumbs":[{"label":"Guides"}]},{"id":"oTgCejs0Z3i8q19kBFgs","title":"Map a dataset instantly","pathname":"/guides/instant-map","siteSpaceId":"sitesp_cwO7v","icon":"sitemap","description":"Drop in CSV, JSON, or connected tables and get back a unified schema with every field mapped, join keys detected, and confidence for each match — no schema, no ETL.","breadcrumbs":[{"label":"Guides"}]},{"id":"2P0FmT4EqEhuumZOxjAJ","title":"Route & execute capabilities","pathname":"/guides/capabilities","siteSpaceId":"sitesp_cwO7v","icon":"diagram-project","description":"Route an objective to the right capability and execute it through the capability plane, using HTTP, the de CLI, or the Python SDK, with execution ownership made explicit.","breadcrumbs":[{"label":"Guides"}]},{"id":"AsjIkFhdc8OenK3Fnqg9","title":"Register capability providers & bindings","pathname":"/guides/capability-bindings","siteSpaceId":"sitesp_cwO7v","icon":"puzzle-piece","description":"Register capability providers, create and test bindings, discover and authorize capabilities, and enable skills — the setup layer behind the Algenta capability plane.","breadcrumbs":[{"label":"Guides"}]},{"id":"vdqrmaPzJkA2Nl0FiSSy","title":"Connect an external MCP server","pathname":"/guides/external-mcp","siteSpaceId":"sitesp_cwO7v","icon":"plug-circle-plus","description":"Register an external MCP server with Algenta — discover its tools into the unified capability catalog and let the engine call them on your agent's behalf, all under the egress policy.","breadcrumbs":[{"label":"Guides"}]},{"id":"t04jAJYsxv6GnFNdndsF","title":"Run an agent","pathname":"/guides/agent-runs","siteSpaceId":"sitesp_cwO7v","icon":"robot","description":"Start a deterministic, replayable agent run through the Algenta Agent Runtime, follow its lifecycle, stream its events, and retrieve its checkpoints and proof artifacts.","breadcrumbs":[{"label":"Guides"}]},{"id":"hSwE3Yg9Tyg5c8xWXEZO","title":"Run an async job","pathname":"/guides/async-jobs","siteSpaceId":"sitesp_cwO7v","icon":"list-check","description":"Submit a long-running simulation as an async job, poll its status by id, fetch the result when it completes, and cancel a queued or running job.","breadcrumbs":[{"label":"Guides"}]},{"id":"hZmJuyqdj0q8UZ8iGpGe","title":"Triage a repository","pathname":"/guides/repository-intelligence","siteSpaceId":"sitesp_cwO7v","icon":"magnifying-glass-chart","description":"Triage a code repository with the Algenta engine — snapshot it, reduce it to a bounded evidence bundle, and read back the findings over the HTTP API","breadcrumbs":[{"label":"Guides"}]},{"id":"Zov8aY5VMsceL76PEBiW","title":"Fix a repository","pathname":"/guides/repository-fix","siteSpaceId":"sitesp_cwO7v","icon":"code-pull-request","description":"Turn a triage evidence bundle into a verified code fix — generate a decision plan, simulate its risk gate, and apply it as a patch, a local branch, or a pull request.","breadcrumbs":[{"label":"Guides"}]},{"id":"5WMmUzHu4Vkt9APe93z3","title":"Prove a security finding","pathname":"/guides/security-analysis","siteSpaceId":"sitesp_cwO7v","icon":"shield-halved","description":"Prove reachability for scanner findings against a repository snapshot, triage by classification, and re-prove that a fix closes the finding after you patch it.","breadcrumbs":[{"label":"Guides"}]},{"id":"uUoN27JEivffHN4OYowW","title":"Register a trigger","pathname":"/guides/triggers","siteSpaceId":"sitesp_cwO7v","icon":"bell","description":"Register a threshold condition on a data source, fire it to run an attached simulation, pause or resume it, and delete it — with optional webhook and auto-execution dispatch.","breadcrumbs":[{"label":"Guides"}]},{"id":"BUaftglZEZi1iiQIpRI6","title":"Webhooks","pathname":"/guides/webhooks","siteSpaceId":"sitesp_cwO7v","icon":"satellite-dish","description":"Receive Algenta events by webhook — async job callbacks, trigger notifications, and decision-execution dispatch — verify delivery with a test POST, and receive inbound GitHub App events.","breadcrumbs":[{"label":"Guides"}]},{"id":"7jxXMejOVXJcCOhtPtta","title":"Bring your own key (BYOK)","pathname":"/guides/byok","siteSpaceId":"sitesp_cwO7v","icon":"key","description":"Supply your own LLM-provider or engine credential. Algenta stores it encrypted at rest, never returns the key material, and uses it only to serve your own org's calls.","breadcrumbs":[{"label":"Guides"}]},{"id":"zzcYa8jjIhdbBp2MBcIs","title":"Device binding","pathname":"/guides/device-binding","siteSpaceId":"sitesp_cwO7v","icon":"fingerprint","description":"Bind an Algenta offline license to a machine identity so it cannot be copied, print the device id, verify a bound license, and fix a device_identity_mismatch.","breadcrumbs":[{"label":"Guides"}]},{"id":"mXMzHBSGgmGVv4p5lT91","title":"Run air-gapped","pathname":"/guides/air-gapped","siteSpaceId":"sitesp_cwO7v","icon":"network-wired","description":"Run Algenta fully air-gapped with no outbound network — install and verify an offline RS256 license, prove the egress policy fails closed, and set the engine for no-LLM or BYOK operation.","breadcrumbs":[{"label":"Guides"}]},{"id":"F3MTK7IWW05ej3r0yE5D","title":"Use Algenta in CI/CD","pathname":"/guides/ci-cd","siteSpaceId":"sitesp_cwO7v","icon":"gears","description":"Run Algenta decisions, simulations, and repository triage from CI/CD with the de CLI — authenticate from a secret, emit JSON, and gate the pipeline on exit codes.","breadcrumbs":[{"label":"Guides"}]},{"id":"vcqdA4B3qUqGzPpMmov0","title":"Run local LLM decode","pathname":"/guides/run-qwen3-decode","siteSpaceId":"sitesp_cwO7v","icon":"gauge-high","description":"Install Algenta, get a supported checkpoint, and run greedy decode on your own Apple Silicon GPU — requirements, limitations, and a complete working example. Currently tested on Qwen3 models.","breadcrumbs":[{"label":"Guides"}]},{"id":"kn3LAnWtPCV6grecFXGD","title":"Model-agnostic LLM inference","pathname":"/guides/run-qwen3-family-inference","siteSpaceId":"sitesp_cwO7v","icon":"gauge-high","description":"Run any supported checkpoint on Apple Silicon through Algenta's model-agnostic SDK path — supported shapes, long context, and the quantized load cache. Currently verified on Qwen3 models.","breadcrumbs":[{"label":"Guides"}]},{"id":"NQEnllpibbqN6mDWhUXj","title":"LLM models & providers","pathname":"/llm-api/models","siteSpaceId":"sitesp_cwO7v","icon":"cubes","description":"The Algenta LLM model catalog — GET /v1/models, the provider.<backend>:<model> id scheme, the ten configurable backends, dynamic model discovery, and per-backend BYOK.","breadcrumbs":[{"label":"LLM API"}]},{"id":"Vn2WOTZlpDBxKz91U7eD","title":"Run chat completions","pathname":"/llm-api/completions","siteSpaceId":"sitesp_cwO7v","icon":"comments","description":"Run chat completions, the unified response envelope, tokenize, and count_tokens over the Algenta LLM API — with streaming, from curl, Python, TypeScript, or the CLI.","breadcrumbs":[{"label":"LLM API"}]},{"id":"rcLVzyXUIqogZji26FoE","title":"Generate embeddings","pathname":"/llm-api/embeddings","siteSpaceId":"sitesp_cwO7v","icon":"chart-scatter","description":"Generate text embeddings, score two vectors with a similarity metric, and rerank candidate documents over the Algenta LLM API — from curl, Python, TypeScript, or the CLI.","breadcrumbs":[{"label":"LLM API"}]},{"id":"9i6iZAg7l3UEawtf7i4b","title":"Python SDK","pathname":"/sdks/python","siteSpaceId":"sitesp_cwO7v","icon":"python","description":"Use Algenta from Python — install algenta-sdk, authenticate, and call the engine with a typed client. Includes runnable examples.","breadcrumbs":[{"label":"SDKs"}]},{"id":"iwYinaOxDZSLtHZNUEcJ","title":"Python client reference","pathname":"/sdks/python/overview","siteSpaceId":"sitesp_cwO7v","icon":"python","description":"Install algenta-sdk, construct the sync or async AlgentaClient, configure retries and timeouts, handle typed errors, and jump to the Python method reference by domain.","breadcrumbs":[{"label":"SDKs"},{"label":"Python SDK","icon":"python"}]},{"id":"l0mqGrhl5ic9OfaGy52Q","title":"Simulations & queries","pathname":"/sdks/python/simulations-and-queries","siteSpaceId":"sitesp_cwO7v","icon":"dice","description":"Python SDK reference for the simulation and query methods — simulate, resolve, query, query_batch, verify, recommend, score, compare, and plan_decision.","breadcrumbs":[{"label":"SDKs"},{"label":"Python SDK","icon":"python"}]},{"id":"azHBRVU9OvmCqePoXOAt","title":"Decisions & products","pathname":"/sdks/python/decisions-and-products","siteSpaceId":"sitesp_cwO7v","icon":"diagram-project","description":"Python SDK reference for decision memory and product intelligence — log_decision, record_outcome, execute_decision, plus product_decision, product_optimize, and product_forecast.","breadcrumbs":[{"label":"SDKs"},{"label":"Python SDK","icon":"python"}]},{"id":"Vhju5AdW90FgzsTAtSnu","title":"Connectors & datasets","pathname":"/sdks/python/connectors-and-data","siteSpaceId":"sitesp_cwO7v","icon":"plug","description":"Python SDK reference for connectors and datasets — create, test, browse, and update connectors, then onboard, list, summarize, refresh, and delete datasets.","breadcrumbs":[{"label":"SDKs"},{"label":"Python SDK","icon":"python"}]},{"id":"eNrZVq5FBOE8nqPiEbrI","title":"Repository intelligence","pathname":"/sdks/python/repositories","siteSpaceId":"sitesp_cwO7v","icon":"code-branch","description":"Python SDK reference for repository intelligence — capabilities, snapshots, triage, decision plans, graph queries, simulate, and apply.","breadcrumbs":[{"label":"SDKs"},{"label":"Python SDK","icon":"python"}]},{"id":"S275RME6M3n4F4QkPcEk","title":"LLM & embeddings","pathname":"/sdks/python/llm","siteSpaceId":"sitesp_cwO7v","icon":"robot","description":"Python SDK reference for the governed LLM surface — list_models, tokenize, count_tokens, chat_completions, responses, embeddings, embedding_similarity, and rerank.","breadcrumbs":[{"label":"SDKs"},{"label":"Python SDK","icon":"python"}]},{"id":"eKSVzJPsfpymqmWCJZvL","title":"Agent runs","pathname":"/sdks/python/agent-runs","siteSpaceId":"sitesp_cwO7v","icon":"robot","description":"Python SDK reference for agent runs — launch, inspect, stream, pause/approve/resume, replay and fork agentic runs, with every AlgentaClient method and its endpoint.","breadcrumbs":[{"label":"SDKs"},{"label":"Python SDK","icon":"python"}]},{"id":"TmAu1XrHTIv0U2a68Jhd","title":"Jobs, triggers & capabilities","pathname":"/sdks/python/jobs-triggers-capabilities","siteSpaceId":"sitesp_cwO7v","icon":"diagram-project","description":"Python SDK reference for async jobs, deployments, triggers, data sources, and the capability plane — every AlgentaClient method with its endpoint and return type.","breadcrumbs":[{"label":"SDKs"},{"label":"Python SDK","icon":"python"}]},{"id":"HOwT1VAS0XihjiBPZnj1","title":"Account & control plane","pathname":"/sdks/python/account-and-control-plane","siteSpaceId":"sitesp_cwO7v","icon":"id-card","description":"Python SDK reference for identity, usage, API keys, teams, billing, devices, audit logs, execution policy, and the platform contract — every AlgentaClient method and endpoint.","breadcrumbs":[{"label":"SDKs"},{"label":"Python SDK","icon":"python"}]},{"id":"OF5vPGpTFlUCu9cANi4Y","title":"TypeScript SDK","pathname":"/sdks/typescript","siteSpaceId":"sitesp_cwO7v","icon":"js","description":"Use Algenta from TypeScript or JavaScript — install the SDK, authenticate, and call the engine with a typed client. Includes runnable examples.","breadcrumbs":[{"label":"SDKs"}]},{"id":"94zbz2BEgat0o5aXo4iU","title":"TypeScript client reference","pathname":"/sdks/typescript/overview","siteSpaceId":"sitesp_cwO7v","icon":"js","description":"Install algenta-sdk, construct an AlgentaClient, and learn the promise-based method surface and the typed error hierarchy before you call the engine.","breadcrumbs":[{"label":"SDKs"},{"label":"TypeScript SDK","icon":"js"}]},{"id":"ajpMQ9mrTXbAhImUrKTA","title":"Simulations & queries","pathname":"/sdks/typescript/simulations-and-queries","siteSpaceId":"sitesp_cwO7v","icon":"chart-line","description":"TypeScript SDK methods for Monte Carlo simulation, decision planning, recommendation, scoring, and the governed query family, each with its HTTP route.","breadcrumbs":[{"label":"SDKs"},{"label":"TypeScript SDK","icon":"js"}]},{"id":"ONhCwMfELeN1mFWHwnXk","title":"Decisions & products","pathname":"/sdks/typescript/decisions-and-products","siteSpaceId":"sitesp_cwO7v","icon":"scale-balanced","description":"TypeScript SDK methods for decision memory (log, list, outcome, execute, delete) and the high-level product endpoints for decision, optimize, retrieve, forecast, and agent runs.","breadcrumbs":[{"label":"SDKs"},{"label":"TypeScript SDK","icon":"js"}]},{"id":"GLLzjTADRlrAr07CtKVA","title":"LLM & model discovery","pathname":"/sdks/typescript/llm","siteSpaceId":"sitesp_cwO7v","icon":"robot","description":"TypeScript SDK methods for the governed LLM surface — chat, responses, embeddings, rerank, tokenization, model artifacts — plus model catalog and runtime discovery.","breadcrumbs":[{"label":"SDKs"},{"label":"TypeScript SDK","icon":"js"}]},{"id":"gq3JggKEFEw0ZB5yhfYz","title":"Connectors & data","pathname":"/sdks/typescript/connectors-and-data","siteSpaceId":"sitesp_cwO7v","icon":"plug","description":"TypeScript SDK methods for connectors, datasets, source registration, and repository intelligence — create, test, browse, onboard, snapshot, triage, and apply.","breadcrumbs":[{"label":"SDKs"},{"label":"TypeScript SDK","icon":"js"}]},{"id":"ccnKaoyvCabCapxXWKrj","title":"Repository intelligence","pathname":"/sdks/typescript/repositories","siteSpaceId":"sitesp_cwO7v","icon":"code-branch","description":"TypeScript SDK reference for repository intelligence — snapshot a repo, triage an issue into ranked evidence, build a decision plan, query the symbol graph, simulate, and apply a fix.","breadcrumbs":[{"label":"SDKs"},{"label":"TypeScript SDK","icon":"js"}]},{"id":"lMeU4kfRqIFAn41wB2FZ","title":"Agent runs","pathname":"/sdks/typescript/agent-runs","siteSpaceId":"sitesp_cwO7v","icon":"robot","description":"TypeScript SDK reference for agent runs — launch, stream, approve, fork, and replay runs, and query their events, checkpoints, mission events, and telemetry.","breadcrumbs":[{"label":"SDKs"},{"label":"TypeScript SDK","icon":"js"}]},{"id":"pGbjtGeuDQ8mo4MvqzUO","title":"Jobs, triggers & capabilities","pathname":"/sdks/typescript/jobs-triggers-capabilities","siteSpaceId":"sitesp_cwO7v","icon":"diagram-project","description":"TypeScript SDK reference for async jobs, webhook triggers, and the capability plane — submit and poll jobs, register triggers, and route, execute, and bind capabilities.","breadcrumbs":[{"label":"SDKs"},{"label":"TypeScript SDK","icon":"js"}]},{"id":"ANUWwL8W8PVMhi2jlWtI","title":"Account & control plane","pathname":"/sdks/typescript/account-and-control-plane","siteSpaceId":"sitesp_cwO7v","icon":"gear","description":"TypeScript SDK reference for the account and control plane — identity, usage, limits, billing, API keys, team, devices, audit logs, execution policy, deployments, health, and version.","breadcrumbs":[{"label":"SDKs"},{"label":"TypeScript SDK","icon":"js"}]},{"id":"9l598Hwbo4aGEvKi96R8","title":"MCP server","pathname":"/sdks/mcp","siteSpaceId":"sitesp_cwO7v","icon":"plug","description":"Connect Algenta to MCP-compatible tools like Cursor and Claude Desktop. Configure the MCP server, authenticate, and verify the tools appear.","breadcrumbs":[{"label":"SDKs"}]},{"id":"paee6eT9BsPydUAeJEsQ","title":"Install from the MCP Registry","pathname":"/sdks/mcp-registry","siteSpaceId":"sitesp_cwO7v","icon":"plug","description":"Install the Algenta MCP server from the Official MCP Registry — io.github.thyn-ai/algenta — into Cursor, Claude Desktop, Claude Code, or Codex, passing your ALGENTA_API_KEY.","breadcrumbs":[{"label":"SDKs"}]},{"id":"MV9BThUlQJmMRg8TA12l","title":"Tool discovery","pathname":"/sdks/tool-discovery","siteSpaceId":"sitesp_cwO7v","icon":"compass","description":"How an AI agent or MCP client discovers Algenta's tools — list the gated tool catalog, read the machine-readable contract, and check the runtime manifest before planning.","breadcrumbs":[{"label":"SDKs"}]},{"id":"apAQ4voB5FWzxcPtVSN4","title":"Testing with Algenta","pathname":"/sdks/testing","siteSpaceId":"sitesp_cwO7v","icon":"vial","description":"Test Algenta-backed code with the engine's determinism — assert stable decision_hash values, replay recorded runs, mock the HTTP transport, and gate live tests on an API key.","breadcrumbs":[{"label":"SDKs"}]},{"id":"UfXsQe4akQMvhWvf2CNv","title":"CLI reference","pathname":"/sdks/cli","siteSpaceId":"sitesp_cwO7v","icon":"terminal","description":"Install and use the supported Algenta command-line surfaces without managing a separate compute runtime.","breadcrumbs":[{"label":"SDKs"}]},{"id":"ZHmoXvADTGBmIZthNjE0","title":"MCP tools overview","pathname":"/mcp-tools/overview","siteSpaceId":"sitesp_cwO7v","icon":"toolbox","description":"The Algenta MCP server exposes the engine as Model Context Protocol tools for Cursor, Claude, and other hosts. Browse the tool catalog by domain.","breadcrumbs":[{"label":"MCP tools"}]},{"id":"acj6XXSxQwONXUAEEH4m","title":"Data & ingest tools","pathname":"/mcp-tools/data","siteSpaceId":"sitesp_cwO7v","icon":"database","description":"MCP reference for Algenta data tools — onboard and train datasets, run the high-level connection flow, register sources, and ingest tables to a simulation.","breadcrumbs":[{"label":"MCP tools"}]},{"id":"s5rO6rro8WVkYO61cVwZ","title":"Connector tools","pathname":"/mcp-tools/connectors","siteSpaceId":"sitesp_cwO7v","icon":"plug","description":"MCP reference for Algenta connector tools — create, inspect, test, browse, update, and delete saved data connectors, plus preview inline definitions.","breadcrumbs":[{"label":"MCP tools"}]},{"id":"tvggF3Vr1L2j4jZIlPKx","title":"Query tools","pathname":"/mcp-tools/query","siteSpaceId":"sitesp_cwO7v","icon":"magnifying-glass","description":"MCP reference for Algenta query tools — governed exact queries, multi-metric batches, and constrained read-only SQL reports over your authorized datasets.","breadcrumbs":[{"label":"MCP tools"}]},{"id":"DCmW2Mt5dbLiqciOLp1G","title":"Repository intelligence tools","pathname":"/mcp-tools/repository-intelligence","siteSpaceId":"sitesp_cwO7v","icon":"code-branch","description":"MCP reference for Algenta's repository-intelligence tools — snapshot, triage, plan, graph-query, simulate, and apply a code fix from any MCP client.","breadcrumbs":[{"label":"MCP tools"}]},{"id":"KbccJY2pq768tDAT4tnS","title":"LLM utility tools","pathname":"/mcp-tools/llm","siteSpaceId":"sitesp_cwO7v","icon":"brain","description":"MCP reference for Algenta's deterministic utility-model tools — list models, tokenize, count tokens, chat, responses, embeddings, similarity, and rerank.","breadcrumbs":[{"label":"MCP tools"}]},{"id":"u2lRG4bQ5F8GKeOwXZYJ","title":"Capability plane tools","pathname":"/mcp-tools/capability-plane","siteSpaceId":"sitesp_cwO7v","icon":"diagram-project","description":"MCP reference for Algenta's capability plane — list providers and bindings, discover and test them, route an objective, execute a capability, and manage skills.","breadcrumbs":[{"label":"MCP tools"}]},{"id":"TArPYdiMnozD75kAybr3","title":"Decision tools","pathname":"/mcp-tools/decisions","siteSpaceId":"sitesp_cwO7v","icon":"scale-balanced","description":"MCP reference for Algenta's decision tools — simulate, recommend, score, compare, the product helpers, and the decision-memory log/outcome loop.","breadcrumbs":[{"label":"MCP tools"}]},{"id":"QxPMt0WNciD57mcVUkS9","title":"Async jobs & triggers tools","pathname":"/mcp-tools/jobs-and-triggers","siteSpaceId":"sitesp_cwO7v","icon":"hourglass-half","description":"MCP reference for async simulation jobs and real-time triggers — submit and poll long-running jobs, test webhooks, and register conditions that auto-run simulations.","breadcrumbs":[{"label":"MCP tools"}]},{"id":"u6tpHpCC255Wd8pXoAfK","title":"Agent run tools","pathname":"/mcp-tools/agent-runs","siteSpaceId":"sitesp_cwO7v","icon":"robot","description":"MCP reference for the Algenta agent-run lifecycle — create runs, read events, checkpoints, mission events and telemetry, and resume, cancel, or approve a run.","breadcrumbs":[{"label":"MCP tools"}]},{"id":"Jop4vFb3jbV8SWK547JZ","title":"Control plane tools","pathname":"/mcp-tools/control-plane","siteSpaceId":"sitesp_cwO7v","icon":"sliders","description":"MCP reference for deployments, team members, devices, audit logs, execution policy, billing, and managed-runtime compatibility.","breadcrumbs":[{"label":"MCP tools"}]},{"id":"IlCUbMTXmhHvhxEW6KJl","title":"Account & meta tools","pathname":"/mcp-tools/account-and-meta","siteSpaceId":"sitesp_cwO7v","icon":"user-gear","description":"MCP reference for account, identity, API keys, usage and runtime discovery — the get_contract, get_me, key-management, run-history and analytics tools.","breadcrumbs":[{"label":"MCP tools"}]},{"id":"0hUbXg4z3RlhTeewoNuK","title":"Algenta for Microsoft Cloud","pathname":"/microsoft-and-azure/overview","siteSpaceId":"sitesp_cwO7v","icon":"microsoft","description":"Turn enterprise data and operating models into governed decisions across Microsoft 365, powered by a decision runtime in your own Azure subscription","breadcrumbs":[{"label":"Microsoft & Azure"}]},{"id":"r20AnmAYgnQnpLBBtqSJ","title":"Decision Agent in Copilot & Teams","pathname":"/microsoft-and-azure/copilot-teams","siteSpaceId":"sitesp_cwO7v","icon":"comments","description":"A declarative Microsoft 365 Copilot and Teams agent that connects to your Algenta runtime over the Model Context Protocol and turns a conversation into a computed decision","breadcrumbs":[{"label":"Microsoft & Azure"}]},{"id":"1PycHcfWzXBuORi0hKwV","title":"Excel Decision Add-in","pathname":"/microsoft-and-azure/excel","siteSpaceId":"sitesp_cwO7v","icon":"file-excel","description":"Run a governed Decision Pack from an Excel task pane over your own decision gateway — selection-only, preview-first, results to new sheets","breadcrumbs":[{"label":"Microsoft & Azure"}]},{"id":"u5JI1wAeSqk4rfOWdu3T","title":"Sign in with Microsoft Entra","pathname":"/microsoft-and-azure/entra-sso","siteSpaceId":"sitesp_cwO7v","icon":"key","description":"Authenticate employees and service identities to the Decision API with audience-scoped Microsoft Entra access tokens, then map each principal to an organization and workspace","breadcrumbs":[{"label":"Microsoft & Azure"}]},{"id":"OqGtUxHNvAjoAGRRoee7","title":"Deploy the decision gateway to your Azure","pathname":"/microsoft-and-azure/gateway","siteSpaceId":"sitesp_cwO7v","icon":"cloud","description":"Deploy the Algenta gateway into your own Azure subscription with a guided Bicep template that keeps the runtime and data plane customer-hosted with cloud features off","breadcrumbs":[{"label":"Microsoft & Azure"}]},{"id":"VPNf9ZSOuh3hdgOdYRej","title":"Sign in with a generic OIDC provider","pathname":"/identity-providers/generic-oidc","siteSpaceId":"sitesp_cwO7v","icon":"key","description":"Authenticate users and service clients to the Decision API with any standards-compliant OIDC provider — Okta, Auth0, Keycloak, or your own — then map each principal to an organization and workspace","breadcrumbs":[{"label":"Identity providers"}]},{"id":"tWtESdNlxR5ZqT7F3czC","title":"Self-hosting","pathname":"/deploy-and-operate/self-hosting","siteSpaceId":"sitesp_cwO7v","icon":"server","description":"Deploy and operate the Algenta engine on your own infrastructure — configure it, run it, and verify a self-hosted instance is healthy.","breadcrumbs":[{"label":"Deploy & Operate"}]},{"id":"7euZkhbiqKc7tFCaSvvP","title":"Configuration reference","pathname":"/deploy-and-operate/configuration","siteSpaceId":"sitesp_cwO7v","icon":"gear","description":"Every Algenta configuration environment variable in one place, with its default and any aliases — complete and current.","breadcrumbs":[{"label":"Deploy & Operate"}]},{"id":"70qQAu9mBa5YatEkZbKV","title":"Monitoring & observability","pathname":"/deploy-and-operate/monitoring","siteSpaceId":"sitesp_cwO7v","icon":"gauge-high","description":"Monitor a self-hosted Algenta deployment — scrape the protected /metrics endpoint, read the engine health signals, tail the structlog JSON logs, all inside your own network.","breadcrumbs":[{"label":"Deploy & Operate"}]},{"id":"b2ly0NyxuqU5iEOtHAFL","title":"Encryption at rest (BYOK)","pathname":"/deploy-and-operate/byok-kms","siteSpaceId":"sitesp_cwO7v","icon":"lock","description":"Encrypt Algenta secrets at rest with your own AES-256-GCM key supplied through ALGENTA_CONFIG_ENCRYPTION_KEY.","breadcrumbs":[{"label":"Deploy & Operate"}]},{"id":"l3PA6TDU1nh826RknZ8y","title":"Device key store","pathname":"/deploy-and-operate/device-key-store","siteSpaceId":"sitesp_cwO7v","icon":"key","description":"Choose where Algenta stores the per-device private key — Apple Secure Enclave, a Linux TPM, the OS keyring, or an encrypted file — and confirm the runtime is healthy with runtime doctor.","breadcrumbs":[{"label":"Deploy & Operate"}]},{"id":"24fPjKQgUryyRxHkd0lS","title":"Audit & tighten egress","pathname":"/deploy-and-operate/privacy-egress","siteSpaceId":"sitesp_cwO7v","icon":"shield-halved","description":"Read the engine's live egress policy, tighten it with deployment-mode settings, generate a 24-hour privacy report, and audit the redacted egress decision log.","breadcrumbs":[{"label":"Deploy & Operate"}]},{"id":"VwKkzV2nlpLhBxMQSY7N","title":"Teams & roles (RBAC)","pathname":"/deploy-and-operate/teams-rbac","siteSpaceId":"sitesp_cwO7v","icon":"users","description":"Invite team members and set roles, understand the owner/admin/member/viewer permission matrix, and read the org audit log and its content-hashed artifacts.","breadcrumbs":[{"label":"Deploy & Operate"}]},{"id":"ElGSv4f85yglNJmlKyMO","title":"Deployments & nodes","pathname":"/deploy-and-operate/deployments-and-nodes","siteSpaceId":"sitesp_cwO7v","icon":"server","description":"Provision an isolated cloud deployment across AWS, Azure, or GCP, track its status and monthly cost, and register self-hosted worker nodes that heartbeat into the engine.","breadcrumbs":[{"label":"Deploy & Operate"}]},{"id":"lPXYo9UbsnGKPrpnU4FQ","title":"Licensing & editions","pathname":"/deploy-and-operate/licensing","siteSpaceId":"sitesp_cwO7v","icon":"scale-balanced","description":"How Algenta licensing works for self-hosted and air-gapped use — what a license grants, how to install and verify it offline, the editions, and the open-core terms.","breadcrumbs":[{"label":"Deploy & Operate"}]},{"id":"LtqjHw8bwZ1rd0qQ3xgl","title":"Usage & limits","pathname":"/deploy-and-operate/billing","siteSpaceId":"sitesp_cwO7v","icon":"gauge-high","description":"Inspect unlimited execution policy and local governance records. Algenta is licensed per device, never per decision.","breadcrumbs":[{"label":"Deploy & Operate"}]},{"id":"4goEB8spNBzblgVDwGQW","title":"Trust overview","pathname":"/trust-and-security/overview","siteSpaceId":"sitesp_cwO7v","icon":"shield-halved","description":"Who is responsible for what across Algenta's deployment modes, and exactly which parties ever see your data.","breadcrumbs":[{"label":"Trust & Security"}]},{"id":"ypLhmyMUFqGAyysTcaoI","title":"Governed execution","pathname":"/trust-and-security/governed-execution","siteSpaceId":"sitesp_cwO7v","icon":"gavel","description":"The control statements behind every plan approval and execution — named denial codes, single-use approval binding, versioned receipts.","breadcrumbs":[{"label":"Trust & Security"}]},{"id":"CEU8v1ihRnhDHN5FYk2Y","title":"Data boundaries","pathname":"/trust-and-security/data-boundaries","siteSpaceId":"sitesp_cwO7v","icon":"diagram-project","description":"Every outbound network call the engine can make is classified into one of five categories and gated by deployment mode — plus what the SSRF guard blocks unconditionally.","breadcrumbs":[{"label":"Trust & Security"}]},{"id":"K1gdi3KQRJ8K3HguTqWR","title":"Audit and evidence","pathname":"/trust-and-security/audit-and-evidence","siteSpaceId":"sitesp_cwO7v","icon":"file-shield","description":"The audit log export carries a content hash you re-derive yourself — evidence you verify, not evidence you take on trust.","breadcrumbs":[{"label":"Trust & Security"}]},{"id":"a51aD1x5TttNAdNBs1Kq","title":"Identity and access","pathname":"/trust-and-security/identity-and-access","siteSpaceId":"sitesp_cwO7v","icon":"id-badge","description":"Four separate identity mechanisms, each scoped to one surface, none falling through to another.","breadcrumbs":[{"label":"Trust & Security"}]},{"id":"uedQpYnZGvlF8QA21MFy","title":"Capability status","pathname":"/trust-and-security/capability-status","siteSpaceId":"sitesp_cwO7v","icon":"list-check","description":"Every protocol surface and integration, labeled Available now, Validated integration, or In conformance testing — checked against the current code and tests, not restated from an announcement.","breadcrumbs":[{"label":"Trust & Security"}]},{"id":"uPY5WAgsssVwHAkVJFfE","title":"Protocol compatibility","pathname":"/trust-and-security/protocol-compatibility","siteSpaceId":"sitesp_cwO7v","icon":"table-columns","description":"MCP version negotiation, the Responses and Chat Completions surfaces, and the 12-scenario conformance suite's real results — 9 passed, 0 failed, 3 blocked with named reasons.","breadcrumbs":[{"label":"Trust & Security"}]},{"id":"vkPEJcNdgitN3TorfAOf","title":"Architecture and trust boundaries","pathname":"/trust-and-security/architecture","siteSpaceId":"sitesp_cwO7v","icon":"sitemap","description":"Where the engine, your model, and your MCP clients run relative to each other — and the one boundary every deployment mode shares.","breadcrumbs":[{"label":"Trust & Security"}]},{"id":"RgDFND9VlVlWosPadOxj","title":"Evidence walkthrough","pathname":"/trust-and-security/evidence-walkthrough","siteSpaceId":"sitesp_cwO7v","icon":"stamp","description":"The versioned execution-receipt envelope, the audited webhook-dispatch path, and the hash-verified audit export — three real artifacts, and an honest note on how they connect today.","breadcrumbs":[{"label":"Trust & Security"}]},{"id":"4qNG32B7GFMLQNbcUzva","title":"Security questionnaire","pathname":"/trust-and-security/security-questionnaire","siteSpaceId":"sitesp_cwO7v","icon":"clipboard-question","description":"Answers to the questions a vendor security review actually asks — each one checked against this repository's code and tests, with the gaps stated as plainly as the controls.","breadcrumbs":[{"label":"Trust & Security"}]},{"id":"dSdAArShiCCFhDje6lJ1","title":"Supply chain","pathname":"/trust-and-security/supply-chain","siteSpaceId":"sitesp_cwO7v","icon":"boxes-stacked","description":"What's verifiable about an Algenta release today — SBOMs and a vulnerability scan per image — and what isn't yet, stated as plainly as what is.","breadcrumbs":[{"label":"Trust & Security"}]},{"id":"WTDdkS9ZmczZYHcwhZmL","title":"Threat model","pathname":"/trust-and-security/threat-model","siteSpaceId":"sitesp_cwO7v","icon":"shield-virus","description":"A public summary of what this engine defends against and how — asset by asset, control by control — without exploit-level detail that would only help an attacker.","breadcrumbs":[{"label":"Trust & Security"}]},{"id":"cTwlkVBklXHphtirwx7a","title":"Support policy","pathname":"/trust-and-security/support-policy","siteSpaceId":"sitesp_cwO7v","icon":"life-ring","description":"What support and long-term-support commitments exist today, and what is explicitly pending an owner decision rather than invented ahead of it.","breadcrumbs":[{"label":"Trust & Security"}]},{"id":"7stbYFWr2YSgMNff3QuV","title":"API overview","pathname":"/http-api/overview","siteSpaceId":"sitesp_cwO7v","icon":"book-open","description":"The Algenta HTTP API — base URL, authentication, conventions, and where to find the full reference and the live explorer.","breadcrumbs":[{"label":"HTTP API"}]},{"id":"dbMexAzWYilfd77tn3O8","title":"API endpoint reference","pathname":"/http-api/reference","siteSpaceId":"sitesp_cwO7v","icon":"code","description":"The complete Algenta HTTP API endpoint reference — every operation grouped by area, with method, path, and summary, generated from the OpenAPI spec.","breadcrumbs":[{"label":"HTTP API"}]},{"id":"bB87ncJpR0SO2QchZ88a","title":"Streaming (SSE)","pathname":"/http-api/streaming","siteSpaceId":"sitesp_cwO7v","icon":"signal-stream","description":"The Algenta Server-Sent Events surfaces — streaming simulation, agent-run event streams, and structured logs — with the exact events, frames, and how to consume them with curl and the SDKs.","breadcrumbs":[{"label":"HTTP API"}]},{"id":"CK7XoiUcZM7OhbPHhEZq","title":"API explorer (live)","pathname":"/http-api/explorer","siteSpaceId":"sitesp_cwO7v","icon":"circle-play","description":"Try the Algenta HTTP API live — set your key and base URL, then send real requests to every endpoint from the browser.","breadcrumbs":[{"label":"HTTP API"}]},{"id":"wwdJd5GQCs0WvhNMBjj7","title":"Error codes","pathname":"/http-api/errors","siteSpaceId":"sitesp_cwO7v","icon":"triangle-exclamation","description":"Every Algenta API error code in one place — generated from the codes the server emits — with what each means and the shape of the error response.","breadcrumbs":[{"label":"HTTP API"}]},{"id":"vujicy2ycDe4OHoN4hSm","title":"API stability","pathname":"/http-api/stability","siteSpaceId":"sitesp_cwO7v","icon":"shield-check","description":"Which Algenta paths carry a compatibility promise, what changes without notice, and which paths sit outside the version scheme on purpose.","breadcrumbs":[{"label":"HTTP API"}]},{"id":"RZaV6MxS14Rjb8HpwvgM","title":"Execution policy & safety limits","pathname":"/http-api/rate-limits","siteSpaceId":"sitesp_cwO7v","icon":"gauge-high","description":"Unlimited Algenta execution, request safety bounds, licensed worker capacity, and configurable abuse controls.","breadcrumbs":[{"label":"HTTP API"}]},{"id":"o0WRO3lUPq7H2rPU17NV","title":"Connectors overview","pathname":"/connectors/overview","siteSpaceId":"sitesp_cwO7v","icon":"plug","description":"Connectors are configured, testable connections to your external data — databases, warehouses, object storage, REST APIs, and files — that back a governed source the engine can query.","breadcrumbs":[{"label":"Connectors"}]},{"id":"UfZPCJCMguJNg9CKFxOe","title":"Connect a database","pathname":"/connectors/databases","siteSpaceId":"sitesp_cwO7v","icon":"database","description":"Connect a database directly to Algenta — the engine reads your tables under a governed contract, never copies them, and keeps credentials encrypted at rest.","breadcrumbs":[{"label":"Connectors"}]},{"id":"lCqcj2JtRPaotaF5GJvm","title":"Connectors reference","pathname":"/connectors/reference","siteSpaceId":"sitesp_cwO7v","icon":"table-list","description":"Every supported Algenta connector type, grouped by category, with its required and optional config fields, the canonical type names, and the untested-live-error status lifecycle.","breadcrumbs":[{"label":"Connectors"}]},{"id":"nYTegW8BuYYDLDwjOQep","title":"Snowflake","pathname":"/connectors/snowflake","siteSpaceId":"sitesp_cwO7v","icon":"snowflake","description":"Create a Snowflake connector with its real config fields, test the connection, browse its tables, onboard one as a dataset, and run a governed query against it.","breadcrumbs":[{"label":"Connectors"}]},{"id":"6eM92gJ0Ohzm1qrddkqr","title":"BigQuery","pathname":"/connectors/bigquery","siteSpaceId":"sitesp_cwO7v","icon":"database","description":"Connect Google BigQuery to Algenta — create a bigquery connector with project and service-account credentials, test it, browse datasets, and run a governed query.","breadcrumbs":[{"label":"Connectors"}]},{"id":"alcWHt4bwGuEa60cUbuI","title":"ClickHouse","pathname":"/connectors/clickhouse","siteSpaceId":"sitesp_cwO7v","icon":"database","description":"Connect ClickHouse to Algenta as a governed connector. Configure host, port, database, user, password, and secure, then test, browse, connect, and query.","breadcrumbs":[{"label":"Connectors"}]},{"id":"gNDPFSP4OWTgmjsmJ3pM","title":"PostgreSQL","pathname":"/connectors/postgres","siteSpaceId":"sitesp_cwO7v","icon":"database","description":"Create a PostgreSQL connector with its real config fields, test the connection, browse its tables, onboard one as a governed dataset, and run a first governed query against it.","breadcrumbs":[{"label":"Connectors"}]},{"id":"snNQKiriXf4vFAdFXlFt","title":"Oracle Database","pathname":"/connectors/oracle","siteSpaceId":"sitesp_cwO7v","icon":"database","description":"Connect Oracle Database or Autonomous Database with Algenta's bundled Thin-mode driver, verify credentials once, browse schemas, and onboard a table or query.","breadcrumbs":[{"label":"Connectors"}]},{"id":"2HWhRiQzJldxPuzHgFLr","title":"MySQL","pathname":"/connectors/mysql","siteSpaceId":"sitesp_cwO7v","icon":"database","description":"Create a MySQL connector with its real config fields, and understand the managed connection-test behavior for MySQL before you onboard data.","breadcrumbs":[{"label":"Connectors"}]},{"id":"1zvEHpqti5c8znv56ZEA","title":"Microsoft SQL Server","pathname":"/connectors/mssql","siteSpaceId":"sitesp_cwO7v","icon":"server","description":"Create a Microsoft SQL Server connector with its real config fields, and understand the managed connection-test behavior for SQL Server before you onboard data.","breadcrumbs":[{"label":"Connectors"}]},{"id":"oDYuQpmAHBPM858ppEjm","title":"SQLite","pathname":"/connectors/sqlite","siteSpaceId":"sitesp_cwO7v","icon":"database","description":"Create a SQLite connector from a database file path, test the connection, browse its tables, onboard one as a governed dataset, and run a first governed query against it.","breadcrumbs":[{"label":"Connectors"}]},{"id":"1x6cl2HefQDDmzc9JPTE","title":"Amazon Redshift","pathname":"/connectors/redshift","siteSpaceId":"sitesp_cwO7v","icon":"aws","description":"Create an Amazon Redshift connector with its real config fields, test the connection over the PostgreSQL wire, browse its tables, onboard one as a governed dataset, and run a first governed query.","breadcrumbs":[{"label":"Connectors"}]},{"id":"Pv08Fr0IvnPwOTKivI9D","title":"Amazon S3","pathname":"/connectors/s3","siteSpaceId":"sitesp_cwO7v","icon":"aws","description":"Create an Amazon S3 connector with its real config fields, test object access, browse a bucket, onboard one object as a governed dataset, and run a first governed query against it.","breadcrumbs":[{"label":"Connectors"}]},{"id":"7daJKOh1ymVXbZvYn7N1","title":"Google Cloud Storage","pathname":"/connectors/gcs","siteSpaceId":"sitesp_cwO7v","icon":"google","description":"Create a Google Cloud Storage connector with its real config fields, test object access, browse a bucket, onboard one object as a governed dataset, and run a first governed query.","breadcrumbs":[{"label":"Connectors"}]},{"id":"mTceQEYGKRUXY86O0KHF","title":"Azure Blob Storage","pathname":"/connectors/azure-blob","siteSpaceId":"sitesp_cwO7v","icon":"microsoft","description":"Create an Azure Blob Storage connector with its real config fields, test blob access, browse a container, onboard one blob as a governed dataset, and run a first governed query.","breadcrumbs":[{"label":"Connectors"}]},{"id":"04t9Go5sMkPDdlVEq9VP","title":"Redis","pathname":"/connectors/redis","siteSpaceId":"sitesp_cwO7v","icon":"bolt","description":"Create a Redis connector with its real config fields, test the connection, browse its keys, onboard selected keys as a dataset, and run a governed query against them.","breadcrumbs":[{"label":"Connectors"}]},{"id":"VtTJSG8QbpgpJiEPxhsU","title":"Neo4j","pathname":"/connectors/neo4j","siteSpaceId":"sitesp_cwO7v","icon":"circle-nodes","description":"Create a Neo4j connector with its real config fields, test the connection, browse its labels and relationship types, onboard a selection as a dataset, and run a governed query against it.","breadcrumbs":[{"label":"Connectors"}]},{"id":"JwmIvw8BUVeUdft7pmJq","title":"Elasticsearch","pathname":"/connectors/elasticsearch","siteSpaceId":"sitesp_cwO7v","icon":"magnifying-glass","description":"Create an Elasticsearch connector with its real config fields, test the connection, browse its indices, onboard one as a dataset, and run a governed query against it.","breadcrumbs":[{"label":"Connectors"}]},{"id":"JYEgoXgC3lW5NAQM1JCW","title":"Files & uploads","pathname":"/connectors/files","siteSpaceId":"sitesp_cwO7v","icon":"file-lines","description":"Create a file connector for CSV, TSV, JSON, Excel, or Parquet data with its real config fields, test that it parses, onboard it as a dataset, and run a governed query against it.","breadcrumbs":[{"label":"Connectors"}]},{"id":"6Ti26sOmZbwwKoJZ5Zjd","title":"Code repositories","pathname":"/connectors/code-repositories","siteSpaceId":"sitesp_cwO7v","icon":"code-branch","description":"Create a repository connector for GitHub, GitLab, Bitbucket, a local checkout, or an archive, test that the source is reachable, and browse its files to feed repository intelligence.","breadcrumbs":[{"label":"Connectors"}]},{"id":"5n12no4k0csK05JYbJfS","title":"REST API","pathname":"/connectors/rest-api","siteSpaceId":"sitesp_cwO7v","icon":"cloud-arrow-down","description":"Create a rest connector for any HTTP JSON API — set base_url and auth_type, test reachability, browse endpoints, connect a dataset, and run a governed query against it.","breadcrumbs":[{"label":"Connectors"}]},{"id":"xhq9AZy4dLzxrmCnM412","title":"Benchmarks","pathname":"/benchmarks/overview","siteSpaceId":"sitesp_cwO7v","icon":"gauge-high","description":"How Algenta measures and publishes performance — what each benchmark claims, the baseline it is measured against, and the rules every published number follows.","breadcrumbs":[{"label":"Benchmarks"}]},{"id":"zQdsSrGjfncQxpUtEuV6","title":"Compute kernels","pathname":"/benchmarks/compute-kernels","siteSpaceId":"sitesp_cwO7v","icon":"gauge-high","description":"Quantized matrix-vector results — the fused NF4 path measured against expand-then-multiply, with correctness asserted against the reference kernel.","breadcrumbs":[{"label":"Benchmarks"}]},{"id":"NmM5SXPmWeWZf5q0lvaO","title":"NVIDIA kernel benchmarks","pathname":"/benchmarks/nvidia-kernel-benchmarks","siteSpaceId":"sitesp_cwO7v","icon":"gauge-high","description":"One portable FP32 kernel, unmodified, measured on NVIDIA T4, L4 and A100 — GEMM and saxpy with bit-exact validation; percentages are against spec-sheet peaks.","breadcrumbs":[{"label":"Benchmarks"}]},{"id":"ojPFGgTgHwpnuNTQjO0v","title":"NVIDIA LLM throughput","pathname":"/benchmarks/nvidia-llm-throughput","siteSpaceId":"sitesp_cwO7v","icon":"gauge-high","description":"Measured LLM inference throughput of the CUDA-native session on NVIDIA A100, H100, H200 and L4 — Qwen3-0.6B, Qwen2.5-7B and Gemma-3-27B prefill and decode against llama.cpp, vLLM, SGLang and TensorRT-","breadcrumbs":[{"label":"Benchmarks"}]},{"id":"I8MOxVoyddh1MXip26Mh","title":"NVIDIA LLM serving","pathname":"/benchmarks/nvidia-llm-serving","siteSpaceId":"sitesp_cwO7v","icon":"server","description":"Measured serving characteristics of the CUDA-native engine on NVIDIA A100, L4 and H100 — concurrent serving, single-stream decode, batched decode, prefill, cold start, VRAM, determinism; Gemma 3 27B T","breadcrumbs":[{"label":"Benchmarks"}]},{"id":"ecYuA6OanB0kmi0jqiIy","title":"Attention","pathname":"/benchmarks/attention","siteSpaceId":"sitesp_cwO7v","icon":"layer-group","description":"Attention kernel results — tiled attention memory against sequence length, rotary embedding cost, and the correctness checks that ran alongside every timing.","breadcrumbs":[{"label":"Benchmarks"}]},{"id":"pZ8UmlE55iZgbQg4DKgc","title":"Normalization","pathname":"/benchmarks/normalization","siteSpaceId":"sitesp_cwO7v","icon":"wave-square","description":"Normalization kernel results — RMSNorm and LayerNorm forward and backward cost, the price of the numerically safe variance, and gradient checks against finite differences.","breadcrumbs":[{"label":"Benchmarks"}]},{"id":"gTi15TNVHkOS7vE77TEy","title":"Feed-forward","pathname":"/benchmarks/feed-forward","siteSpaceId":"sitesp_cwO7v","icon":"diagram-project","description":"Feed-forward and embedding kernel results — gated activation cost across three variants, the two GELU conventions and why they are not interchangeable, and embedding scatter cost.","breadcrumbs":[{"label":"Benchmarks"}]},{"id":"Jp5FCaDc1gk3wP5wnxYT","title":"Code retrieval quality","pathname":"/benchmarks/code-retrieval","siteSpaceId":"sitesp_cwO7v","icon":"magnifying-glass-arrow-right","description":"Code retrieval quality on CoIR, measured with nDCG@10 against BM25 baselines evaluated on the same harness.","breadcrumbs":[{"label":"Benchmarks"}]},{"id":"RL6ta4G4WAAXAqStoRF4","title":"Dataframe group-by","pathname":"/benchmarks/dataframe-groupby","siteSpaceId":"sitesp_cwO7v","icon":"table-cells","description":"Group-by aggregation on the H2O db-benchmark, 0.5 GB tier, through Algenta's composed route — six of six questions ahead of both DuckDB 1.5.5 and Polars 1.43.1 on the minimum.","breadcrumbs":[{"label":"Benchmarks"}]},{"id":"50f986k9xaEMM8LwvHii","title":"Sort","pathname":"/benchmarks/sort","siteSpaceId":"sitesp_cwO7v","icon":"arrow-down-1-9","description":"LSD radix sort over Float64 keys against numpy.sort's introsort — 12 of 14 gate-clean cells ahead across three input orders, from 100,000 to 102.4 million elements.","breadcrumbs":[{"label":"Benchmarks"}]},{"id":"x05wUlZ2MpuAix4uAMDv","title":"Top-k selection","pathname":"/benchmarks/top-k","siteSpaceId":"sitesp_cwO7v","icon":"filter","description":"Bounded top-k selection over a flat Float64 column, against numpy.argpartition — a real loss at the smallest size tested, and a 5x-163x win at every size above it.","breadcrumbs":[{"label":"Benchmarks"}]},{"id":"nhunAe8rpfCaJkQJWxaa","title":"Graph betweenness centrality","pathname":"/benchmarks/graph-betweenness","siteSpaceId":"sitesp_cwO7v","icon":"diagram-project","description":"Brandes betweenness centrality over a CSR adjacency, against networkx — ahead in every one of 6 cells, from 3.0x at 200 nodes to 4.3x at 2,000.","breadcrumbs":[{"label":"Benchmarks"}]},{"id":"xyfpxewuy1juNbUpWhrB","title":"Survival analysis","pathname":"/benchmarks/survival-analysis","siteSpaceId":"sitesp_cwO7v","icon":"heart-pulse","description":"Kaplan-Meier and Cox proportional hazards, against lifelines 0.30.0, on two current real-world datasets plus synthetic scaling — wins from 3.0x to 142.9x, one loss disclosed.","breadcrumbs":[{"label":"Benchmarks"}]},{"id":"PBmXvOYaUdNd9FMGpZjD","title":"Least-squares accuracy","pathname":"/benchmarks/least-squares-accuracy","siteSpaceId":"sitesp_cwO7v","icon":"bullseye","description":"Least-squares accuracy on the NIST StRD linear-regression suite, measured in correct digits against the certified values, against fourteen reference arms on the same data.","breadcrumbs":[{"label":"Benchmarks"}]},{"id":"sv1cy3rOpZNC1GXUJ36D","title":"Forecasting on the joint axis","pathname":"/benchmarks/forecasting-joint-axis","siteSpaceId":"sitesp_cwO7v","icon":"stopwatch","description":"Forecast accuracy and latency measured together on all 100,000 M4 series against statsforecast 2.1.1 — equal or better accuracy at 2.42x-3.99x the speed.","breadcrumbs":[{"label":"Benchmarks"}]},{"id":"EuY2c3xxBMAntVwA2Qz8","title":"FinQA exact execution","pathname":"/benchmarks/finqa-exact-execution","siteSpaceId":"sitesp_cwO7v","icon":"calculator","description":"Execution accuracy on all 8,281 gold programs in the real FinQA dataset — Algenta's exact-arithmetic driver against three naive-reimplementation baselines, on the dataset's own exact-match rule.","breadcrumbs":[{"label":"Benchmarks"}]},{"id":"xaVbhLD1TVrIAp719yLQ","title":"LLM inference overview","pathname":"/benchmarks/llm-inference-overview","siteSpaceId":"sitesp_cwO7v","icon":"gauge-high","description":"LLM inference on Apple silicon — prefill, single-stream decode, and batched decode for Qwen3-0.6B-Base, measured against llama.cpp and mlx-lm.","breadcrumbs":[{"label":"Benchmarks"}]},{"id":"Yq9Ne2itDGfgjmW08TSP","title":"Prefill (token input) throughput","pathname":"/benchmarks/llm-prefill-throughput","siteSpaceId":"sitesp_cwO7v","icon":"arrow-right-to-bracket","description":"Prompt prefill (token-input) throughput for Qwen3-0.6B-Base on Apple silicon — a 256-token prompt at ~10,700 tok/s, ahead of llama.cpp's ~10,199, on any prompt length.","breadcrumbs":[{"label":"Benchmarks"}]},{"id":"mSGCsrqc462rjkTu7gGN","title":"Decode throughput","pathname":"/benchmarks/decode-throughput","siteSpaceId":"sitesp_cwO7v","icon":"bolt","description":"Decode tokens per second for Qwen3-0.6B-Base on Apple silicon — 324.8 tok/s median vs mlx-lm 281.6 and llama.cpp 254.0, interleaved order-rotated rounds on the same machine.","breadcrumbs":[{"label":"Benchmarks"}]},{"id":"osRlBh1k3XF8QR2OVDHI","title":"Batched decode throughput","pathname":"/benchmarks/batched-decode-throughput","siteSpaceId":"sitesp_cwO7v","icon":"layer-group","description":"Batched (continuous-batching) decode for Qwen3-0.6B-Base on Apple silicon — many sequences per weight sweep. 2.1-2.6x llama.cpp concurrent throughput at B=8-32, token-verified.","breadcrumbs":[{"label":"Benchmarks"}]},{"id":"IthHAvoSl3XlWPgOhILY","title":"LLM tool-calling uplift","pathname":"/benchmarks/llm-tool-calling-uplift","siteSpaceId":"sitesp_cwO7v","icon":"robot","description":"gpt-5-mini answering 16 real questions against a 33,125-row CSV, with and without Algenta's tool-calling access to its own compiled engine — 16 of 16 correct with tools, 1 of 16 without.","breadcrumbs":[{"label":"Benchmarks"}]},{"id":"RaQHf2GXLycvWnyQW0hI","title":"Troubleshooting","pathname":"/help/troubleshooting","siteSpaceId":"sitesp_cwO7v","icon":"life-ring","description":"Hit a snag? Find your message, copy the fix, get back to work — most Algenta issues are a missing key, an engine it can't reach, or a license flag.","breadcrumbs":[{"label":"Help"}]},{"id":"bdxaaDSHzNrLgMm0q7zu","title":"Glossary","pathname":"/help/glossary","siteSpaceId":"sitesp_cwO7v","icon":"book-bookmark","description":"Every Algenta term in one place — the engine, the surfaces, decision planning, governed data, agentic runs, deployment, and licensing vocabulary, each defined once.","breadcrumbs":[{"label":"Help"}]},{"id":"mUN6X4yj3DJmXqbTs2Kv","title":"Quick reference","pathname":"/help/quick-reference","siteSpaceId":"sitesp_cwO7v","icon":"table-list","description":"One-page cheat sheet — every common Algenta task mapped to its HTTP endpoint, de CLI command, and Python SDK call, plus the handful of env vars you need most.","breadcrumbs":[{"label":"Help"}]},{"id":"SXfuMhUztqRBzvLqAX2w","title":"Using these docs with AI","pathname":"/help/ai-access","siteSpaceId":"sitesp_cwO7v","icon":"robot","description":"Use the Algenta docs with any AI — the auto-generated llms.txt context files, per-page Markdown, the built-in MCP server, and copy/open-in-chat actions.","breadcrumbs":[{"label":"Help"}]},{"id":"UBadLVsBUFTqjIUvjThP","title":"Changelog","pathname":"/help/changelog","siteSpaceId":"sitesp_cwO7v","icon":"clock-rotate-left","description":"What's new in Algenta — notable releases, improvements, and fixes to the engine, SDKs, MCP server, and self-hosted deployment, newest first.","breadcrumbs":[{"label":"Help"}]}]}