Practical skill tracks for people building real AI systems.
Learn the execution skills around prompts, evaluation, agents, and workflow design without drifting into theory-only content. These tracks are meant to feed directly into the product surfaces where the work actually happens.
Use the track pages as a bridge into real practice.
Read just enough to understand the loop, then move into prompts, challenges, or workspace flows while the concept is still fresh.
Featured tracks
High-signal starting points for builders leveling up around AI systems.
Build disciplined prompt structure, evaluation habits, and iteration loops that hold up in production instead of only in demos.
Learn how to define good outputs, compare variants, and use feedback loops instead of one-off subjective judgment.
Use async coding agents, repo context, and challenge workflows together so AI output stays grounded in real tasks.
Work from the CLI when you want to browse, start, and submit from a repo-first workflow instead of clicking through the app.
Pair these with
What a strong skill loop looks like
Learn the concept from docs or a guide.
Inspect examples from prompt, challenge, or workspace surfaces.
Practice against a real task, then compare variants instead of trusting a single run.