// the find
NirDiamant/Prompt_Engineering
22 prompt engineering techniques with hands-on Jupyter Notebook tutorials, from fundamental concepts to advanced strategies for leveraging LLMs.
A collection of 22 Jupyter notebooks walking through prompt engineering techniques, from zero-shot and few-shot basics up to tree-of-thought and self-consistency. Aimed at developers who want hands-on examples rather than just theory. The companion to the author's RAG Techniques and GenAI Agents repos.
The progression from basic prompt structures through optimization and evaluation is genuinely well-sequenced — you can follow it linearly without hitting gaps. The security and safety notebook covering prompt injection is a topic most tutorial repos skip entirely. Coverage of self-consistency (sampling multiple reasoning paths and aggregating) is explained practically, not just theoretically. The repo is actively maintained with recent commits and has a real community around it.
Everything goes through OpenAI or LangChain — there is no treatment of how these techniques behave differently across model families, which matters a lot in practice. The 'evaluation' notebook is the weakest: it covers metric setup but doesn't engage with the hard problem of automated prompt evaluation at scale. The non-commercial license is a quiet gotcha for anyone wanting to use these in a production training pipeline or internal tooling. The repo is essentially a funnel for the author's paid course and newsletter, which colors how some content is scoped.