Resources
Companion materials
Three resources referenced throughout the course. They're meant to be working tools, not just reading material — copy what you need into your own projects.
Prompt library
A curated set of vetted prompts for common tasks: classification, extraction, summarization, generation, RAG, reasoning, and meta-prompts. Each prompt is annotated with what it's for, why it's structured the way it is, and where to tweak it. Click-to-copy on every block.
Eval starter kit
A working eval harness in Python: dataset format, three reference graders (exact match, regex, LLM-as-judge), a runner that emits diffs against a baseline, and a GitHub Actions workflow for CI. About 150 lines of code, designed to be copied into your repo and adapted.
Cheat sheet
A one-page summary of the patterns from the course and when to reach for each: the prompt skeleton, technique selection, sampling parameters, failure-mode lookup, latency levers, eval discipline, injection defenses, and a pre-launch checklist. Designed to print cleanly as PDF.
Haven't taken the course yet?
These resources make a lot more sense in context. The course explains why these patterns work and when each one is the wrong choice.
Learn more →