Research, software, data and methods. Inspect the context, compare your options, and take a useful next step.
Upstream project repository
Evaluation tooling for retrieval-augmented and language-model applications, including metric-driven comparison workflows.
Investigate How sensitive is one evaluator to controlled answer perturbations on a tiny authored retrieval fixture?
Upstream project repository
Raster data access, georeferencing, and related operations through GDAL.
Investigate Do window reads and pixel-to-world mappings agree with a known synthetic raster and affine transform?
Upstream project repository
Distributed execution and machine-learning infrastructure for studying task scheduling, tuning and fault handling.
Investigate Does a small synthetic parameter sweep retain each trial's identity and failure state under a declared retry policy?
Upstream project repository
N-body simulation with gravitational dynamics integrators and analysis interfaces.
Investigate How do selected time steps affect energy and orbital-phase error for an analytic two-body orbit?
Upstream project repository
Planning and execution of array rechunking under an explicit memory budget.
Investigate Does a bounded rechunking plan preserve every cell while respecting its declared intermediate-store design?
Upstream project repository
Reprojection of astronomical images between coordinate grids.
Investigate How do selected reprojection methods affect a constant field and an isolated normalized source?
Upstream project repository
Geospatial raster operations and metadata access for xarray objects.
Investigate Does a simple rectangular clip preserve expected pixels, CRS, and transform metadata?
Upstream project repository
Incremental machine learning and evaluation for data streams.
Investigate How quickly does a fixed online learner adapt after a declared change in the generating process?
Upstream project repository
Experiment organization and observation tooling for capturing configurations, dependencies and run outcomes.
Investigate Does a controlled failed run preserve its configuration and failure outcome as clearly as a successful run?
Upstream project repository
Sparse-autoencoder research tooling for examining reconstruction, sparsity and feature interventions on neural activations.
Investigate How does a fixed sparsity setting trade reconstruction error against active features on an authored low-dimensional activation set?
Upstream project repository
Tensor serialization tooling for studying shape, dtype and byte-level preservation without a pickle-style object format.
Investigate Do small authored tensors round-trip with exact dtype, shape and payload bytes through the selected serializer?
Upstream project repository
Sampling and analysis methods for global sensitivity analysis of computational models.
Investigate How do estimated variance-based sensitivity indices converge for a fixed analytic benchmark?
Recorded license evidence is scoped to each source; it is not a blanket permission or a verification result.