We’re starting Module 2: Vector Search

We’ve updated and expanded it together with Qdrant.

It now covers:

- Getting started with Qdrant (vector DB)
- Text embeddings
- Indexing and search
- RAG with vector search
- Hybrid search techniques
🔥 New hands-on notebooks

Here’s what you need to do to complete this module:

1. Go through the module content
2. Watch the videos and explore the notebooks
3. Set up Qdrant locally (optional, but recommended)
4. Do the homework
5. Submit the homework here

Deadline: 1 July, 1 AM CET (in 2 weeks)

Good luck and enjoy Module 2!

We’ve prepared a special workshop for you as part of the LLM Zoomcamp!

Join us for a live hands-on session to learn how to build a smart, always-up-to-date data pipeline for your RAG projects.

LLM Zoomcamp students: you’ll earn bonus points on the leaderboard if you submit the homework for this workshop.

Date: Monday, June 30
Time: 4:30 PM CET

In this workshop, you’ll learn how to:

- Extract data from REST APIs using dlt
- Vectorize and load it into a database
- Index it semantically with Cognee
- Handle dynamic, evolving relationships between records
- Connect everything into a portable, open-source pipeline you can run anywhere: from notebooks to Airflow, Dagster, or Mage

By the end, you’ll have a fully working ingestion pipeline that powers RAG built entirely with open-source tools.

Register here: https://lu.ma/kdp6ex5s

Looking forward to building with you!

Tomorrow (Tuesday) we will have office hours where we will talk about vector search

Jenny and Kacper will join us for the stream

Ask your questions here: https://app.sli.do/event/hXoPWt27hc1QZJdArvX2MQ

Time: 17:00 Berlin time

See you tomorrow! And if you can't make it, don't worry, there will be a recording

You can join our Open-Source LLM pre-course QnA session

Stream: https://www.youtube.com/watch?v=mmUyeyAPVnU

Questions: https://app.sli.do/event/rTvJifGJFiMWadwwfnar9J

We're starting in a couple of minutes

Stream: https://www.youtube.com/watch?v=7pHsGGucJ4I
Questions: https://app.sli.do/event/hXoPWt27hc1QZJdArvX2MQ

Schedule update

Modules 3 and 4 need a bit more polishing, so we’re pushing them back.

To keep the momentum, we’re adding a bonus livestream next week!

Date: Tuesday, 1 July
Time: 4:30 PM CET

Alexey will do a hands-on workshop on turning RAG pipelines into agentic AI flows.

What we’ll cover
- Build a basic RAG pipeline (search + prompt + LLM)
- Add agentic behaviour: decide when to search vs. answer
- Run multi-step, reasoning-driven searches
- Use OpenAI function calling and chat_assistant.py
- Streamline tools with PydanticAI

You’ll get a working assistant that
- searches the FAQ
- decides when to act
- asks follow-up questions
- updates its own knowledge base

Register here: https://lu.ma/b7p9p365

We're starting a workshop in 30 minutes. Be prepared!

We will share the link here once we start

See you soon

Workshop: https://www.youtube.com/watch?v=MNt_KK32gys
Notebook: https://colab.research.google.com/drive/1vBA9OIGChcKjjg8r5hHduR0v3A5D6rmH?usp=sharing

We’re starting our workshop about agents in 15 minutes!

We’ll share the link here as soon as we go live.

Get ready, we begin shortly. See you soon!

Stream: https://www.youtube.com/watch?v=GH3lrOsU3AU
Workshop repo: https://github.com/alexeygrigorev/rag-agents-workshop
Course repo: https://github.com/DataTalksClub/llm-zoomcamp/tree/main/0a-agents

The homework for the dlt workshop is ready!

In this homework, you will ingest FAQ data with dlt into Qdrant

Here it is: https://github.com/DataTalksClub/llm-zoomcamp/blob/main/cohorts/2025/workshops/dlt.md

Have fun and enjoy learning!

The Agents module is now fully ready! We have added homework, in which you will learn about MCP

Module: https://github.com/DataTalksClub/llm-zoomcamp/tree/main/0a-agents
Homework: https://github.com/DataTalksClub/llm-zoomcamp/blob/main/cohorts/2025/0a-agents/homework.md

Have fun!

For this cohort of LLM Zoomcamp we have prepared something special for you!

Instead of a competition, like last year, we're hosting a hackathon!

In this Hackathon you can put the skills you picked up during the course to practice and make something useful for the community

Here's more information about it: https://github.com/DataTalksClub/llm-zoomcamp/tree/main/cohorts/2025/hackathon

You can submit your ideas here: https://airtable.com/appjxMzaGej3KHwEh/shr12Sx3SGsStvjEd

And we will discuss them on the launch stream next week

https://lu.ma/i98j574k

See you next week!

We’re starting Module 3: Evaluation

This module walks through how to assess search and RAG systems, both offline and online, using ground-truth data and an LLM as judge.

What’s covered:

- Why evaluation matters
- Search evaluation
- Ranking evaluation (text search)
- RAG evaluation

Watch the content here:

- Module content
- Course playlist

Homework:

- Link
- Deadline: Tuesday 22 July 2025, 23:59 CET
- Submit here

We’re starting the Summer 2025 AI Hackathon with Alexey’s live launch session today!

This hackathon is part of the LLM Zoomcamp where you’ll build LLM-powered tools for the DataTalks.Club community this summer.

Alexey will explain:
- Hackathon goals and structure
- Idea submission and selection criteria
- Project examples
- Team formation
Plus a live Q&A

Next steps:

- Join #hackathon on Slack (go solo or form teams of up to 4)
- Review suggested ideas and submit your own before the stream for live feedback
- Register for the launch session with Alexey

See you at the launch!

We're starting our hackathon stream in 1 hour!

Ask your questions here: https://app.sli.do/event/hXoPWt27hc1QZJdArvX2MQ

Submit your ideas for discussion here: https://airtable.com/appjxMzaGej3KHwEh/shr12Sx3SGsStvjEd

We will share the link to the live stream 5-10 minutes before the start. See you soon!

Hackathon launch stream!

Watch here: https://www.youtube.com/watch?v=APtJ1SDGdw0
Ask questions: https://app.sli.do/event/hXoPWt27hc1QZJdArvX2MQ

Available for replay later

For module 3 this year, we used materials from multiple modules of 2024 edition

Since it might be confusing at times, here's a video that explains the changes

https://www.youtube.com/watch?v=qiWmCV0Knl0&list=PL3MmuxUbc_hIB4fSqLy_0AfTjVLpgjV3R

It's optional - if you already watched the module and didn't get confused, you don't need it

Hi all!

We're still preparing the monitoring module, so we're extending the deadline for module 3 homework by one week

We will also keep the agents homework open in case you didn't have a chance to work on it