AIDevWorks: AI Architect Track – Vector Databases Deep Dive

What you’ll learn (session focus)

• Vector DBs & embeddings: what they are, when to use them
• Architecture overview: Pgvector, Redis Vector, ChromaDB, Pinecone
• Real-world setup: indexing, metadata
• LLM connection: end-to-end retrieval flow and prompt handoff

Hands-on walkthrough

• Step-by-step setup of one or more vector DBs
• Create an embedding pipeline (documents → chunks → vectors)
• Implement a retrieval + response chain with LangChain / Ollama