Probabl Skills

Skills

14 in the Claude listing

• Audit ml pipelineOwns the `audit/` folder: one `# %%` (jupytext percent) Python file per experiment, aligned 1:1 with `experiments/NN_<short_name>.py` and `journal/NN_<short_name>.md`, that loads the experiment's skor
• Build ml pipelineDeclare the pipeline from data source to predictor as a skrub DataOps graph (not as a bare `sklearn.Pipeline`).
• Data science python stackOpinionated Python stack for data-science / ML work — one library per job, organized into tiers (mandatory / user choice / optional / transitive).
• Evaluate ml pipelineMethodology for evaluating a single sklearn-compatible learner (in particular, the `SkrubLearner` produced by `build-ml-pipeline`).
• Explore ml dataOwns data understanding BEFORE any model is designed.
• Iterate from skoreSource the next ML experiment proposal by reading the audit digest at `scratch/audit/<stem>/audit.md` (produced by `audit-ml-pipeline` at § 4 record-outcome).
• Iterate from userSource the next ML experiment proposal from the user via one of three entry points selected by `AskUserQuestion`: (a) a scientific article URL the agent must read and synthesize, (b) a resource link o
• Iterate ml experimentOwns the iteration loop on top of an ML workspace: the `journal/JOURNAL.md` index and the per-experiment `journal/NN_short_name.md` design notes that must be drafted and approved by the user before*
• Organize ml workspaceDecide where files live in an ML experimentation project: reusable code in `src/<pkg>/`, one `# %%` script per experiment in `experiments/`, design notes + index in `journal/`, reports in `reports/`,
• Python apiLook up the public API of a Python package against the *installed version* and cache what's worth keeping.
• Python code styleOwns Python code style for this stack: ruff for lint + format, numpydoc for docstrings.
• Python env managerSingle source of truth for "which Python environment manager does this project use, and how do I install a package with it?".
• Smoke test ml pipelineOwns the smoke test contract for an ML experiment: a small, diagnostic-by-construction pytest that fits the experiment's learner on a portion of the real `data/` source and predicts on a *disjoint* po
• Test ml pipelineOwns the `tests/` folder of an ML workspace and the pairing rule between an experiment and its tests.