LLM Zoomcamp
The LLM Zoomcamp is a free, hands-on course on building real-world applications with Large Language Models. Each cohort runs once a year and focuses on AI engineering: using hosted LLMs (OpenAI, Groq, and others) to build production-style RAG (retrieval-augmented generation) systems, evaluate them, and monitor them in production.
The 2026 edition runs from June through the end of summer. About 10 weeks: roughly two weeks for module 1 and one week each for the remaining modules, plus three weeks for the final project.
LLMs themselves are treated as black boxes here. The course is for people who want to build with LLMs rather than train them.
How to Use These Docs
Read in this order:
- Community Guidelines - code of conduct, how to ask questions, how to use Slack channels, how to promote your work.
- Zoomcamp Logistics - how DataTalks.Club zoomcamps work in general (cohort schedule, joining, live sessions, homework, project, peer review, certification). Most of your logistical questions are answered there.
- The pages in this section - what is specific to the LLM Zoomcamp (curriculum, environment, project rubric, LLM-specific resources).
- The LLM Zoomcamp FAQ - module-specific and technical questions from previous cohorts.
For platform mechanics (where to click on the submission form, how the leaderboard appears), see Course Management Platform.