Repolode
Submit
DataTalksClub

DataTalksClub/llm-zoomcamp

A free, open-source 10-week online course on building LLM applications with RAG, vector search, agents, evaluation and monitoring.

View on GitHubWebsite
Stars
7,493
Forks
1,391
Open issues
6
Last commit
5 days ago

What is it?

LLM Zoomcamp is a free, hands-on online course run by DataTalks.Club. It walks through building an AI system that answers questions over your own knowledge base, covering retrieval-augmented generation (RAG), embeddings, vector search, function calling and AI agents.

The course is organized into modules on agentic RAG with keyword search, vector search with minsearch, sqlitesearch and PGVector, orchestration with Kestra, data ingestion with dlt, evaluation, monitoring, and best practices such as hybrid search and reranking. Later modules include an end-to-end fitness assistant example.

The repository holds the course materials, notebooks and homework. Students can take it in a live cohort with graded homework, a capstone project and peer review, or self-paced. Prerequisites are Python, the command line and basic Docker; no GPU is needed, and running the exercises costs roughly a few dollars in API credits.

Key features

  • 01Covers RAG pipelines, embeddings, vector search and hybrid search
  • 02Includes modules on function calling and AI agents
  • 03Compares search backends such as minsearch, sqlitesearch and PGVector
  • 04Teaches evaluation and monitoring of LLM applications
  • 05Ends with a capstone project and optional peer review in live cohorts

Use cases

  • Software engineers adding RAG-based search or Q&A to an existing product
  • Data engineers learning how retrieval pipelines fit into production systems
  • ML practitioners who want a structured approach to evaluating LLM apps

Things to know

  • No license is specified in the repository metadata, so reuse terms are unclear.
  • Running the exercises costs about $1-5 in OpenAI API credits.
  • Certificates and graded homework require joining a live cohort with deadlines.

Topics

#llm#rag#vector-search#embeddings#ai-agents#course#python

FAQ

What is llm-zoomcamp?
A free, open-source 10-week online course on building LLM applications with RAG, vector search, agents, evaluation and monitoring. LLM Zoomcamp is a free, hands-on online course run by DataTalks.Club. It walks through building an AI system that answers questions over your own knowledge base, covering retrieval-augmented generation (RAG), embeddings, vector search, function calling and AI agents.
Is llm-zoomcamp free to use? What license does it have?
llm-zoomcamp has no declared license. Its code is public, but without a license you have no explicit permission to reuse it; ask the author before relying on it.
Is llm-zoomcamp still maintained?
Actively maintained. The last commit to llm-zoomcamp was on Oct 15, 2026.
What can I use llm-zoomcamp for?
Software engineers adding RAG-based search or Q&A to an existing product Data engineers learning how retrieval pipelines fit into production systems ML practitioners who want a structured approach to evaluating LLM apps
Where can I get llm-zoomcamp?
The source code and installation instructions are on GitHub: https://github.com/DataTalksClub/llm-zoomcamp

Summary generated by AI from the project's README. GitHub stats refresh daily.

Related projects