Build Your Own AI-Powered Document Search

​See our site for full details

​In this hands-on workshop, you'll build a working RAG system from scratch. You'll ingest documents, generate embeddings, store them in a vector database, write similarity queries, and wire everything into a chat interface. By the end, you'll have a functional AI-powered document search running locally—and understand every piece of the pipeline.

​A hands-on workshop where you'll build a working RAG system from scratch. Not slides, not theory, a system you built.

​In this on-site 2 hour workshop (Instruction time 12:30-2:30pm), you will learn:
- Understand why RAG exists and how it solves LLM hallucination and knowledge problems by grounding responses in your own documents.
- Build a complete ingestion pipeline that chunks documents, generates embeddings, and stores vectors in ChromaDB.
- Query your vector database using similarity search and interpret results with tuning parameters like top-k and thresholds.
- Connect retrieval to a web chat interface, completing the full RAG loop from question to grounded answer.

​This is limited seating so please only RSVP if you are sure you can make it.

​Parking: Closest Garage I see is: https://maps.app.goo.gl/Z4z3Ln6fSVhK8U589
On street:
https://pdx.maps.arcgis.com/apps/MapSeries/index.html?appid=ad171d005d4442bba3c640735d070aa3