with one click
oreilly-mlops-with-databricks
oreilly-mlops-with-databricks contains 6 collected skills from mvechtomova, with repository-level occupation coverage and site-owned skill detail pages.
Skills in this repository
Evaluate a student MLOps/LLMOps repository against the course best-practices checklist and produce a scored report. Trigger whenever the user asks to evaluate, grade, review, or score a student repo (or any external repo) against the best practices, the checklist, or the course rubric, e.g. "evaluate ~/repos/student-x", "grade this repo", "check whether this repo follows the best practices".
Create an Excalidraw JSON diagram in visuals/ from a written description (no source image), using the project's brand palette and styling. Trigger whenever the user describes a diagram in words and asks to draw/create/make it in Excalidraw, sketch a diagram from a spec, build a new diagram from scratch, or otherwise wants an Excalidraw file without an existing PNG to convert.
Convert raw PNG diagrams in visuals_raw/ into Excalidraw JSON files in visuals/ that match the brand palette and styling. Trigger whenever the user asks to turn a PNG (or screenshot/sketch) into an Excalidraw diagram, recreate a diagram in Excalidraw, restyle a PNG diagram with the project palette, or mentions converting files under visuals_raw/.
Deploy and run a Databricks notebook via Databricks Asset Bundles (DABs). Trigger whenever the user asks to run, deploy, or execute a notebook on Databricks, run a .py notebook from llmops/ or mlops/, kick off a Databricks job for a notebook, or set up a bundle job resource for an existing notebook.
Update every dependency in pyproject.toml to its latest stable version from PyPI and resync the environment with uv. Trigger whenever the user asks to fix/update/bump/refresh dependencies, sync deps to latest, upgrade pyproject.toml versions, or pin packages to the newest stable releases.
Commit all staged and unstaged changes on the current branch, then push to the remote. Trigger whenever the user says "ship", "ship it", asks to commit and push, push the branch, finalize a feature branch, or otherwise wants outstanding local changes committed and pushed in one step. Refuses to operate directly on main/master.