Joon-Pil Hwang

Weaviate

- Introduction to vector databases & Weaviate
   - Overview of vector embeddings
   - How to find the right data - keyword (BM25), vector and hybrid searches + filtering
   - Motivation for retrieval augmented generation (RAG)
   - Hands-on
       - Set up Weaviate & perform data ingestion
       - Perform searches & RAG


- Vector databases in depth
   - Indexing techniques to speed up search (HNSW & inverted indexes)
   - Improving efficiency with quantization / encoding techniques
   - Working with embedding models & LLMs, using Weaviate
   - Architectural considerations (multi-tenancy implementations)
   - Hands-on
       - Work with larger datasets
       - Try different embedding/language models
       - Build a multi-tenant database


- Next steps with a vector database
   - Multi-modal data and Weaviate
   - Building an agentic workflow with Weaviate
   - Introduction to Weaviate Agents
   - Hands-on:
       - Work with multi-modal embeddings
       - Convenient interactions with Weaviate agents