Level 2 · Intermediate Production DBs

Goal: move past toy examples and get hands-on with the vector databases teams actually run in production — managed and self-hosted — plus the indexing theory behind why they're fast.

Modules

1. Pinecone Fundamentals
2. Weaviate: Schemas & Hybrid Search
3. Qdrant: Collections & Payloads
4. pgvector: Vectors Inside PostgreSQL
5. How HNSW Actually Works
6. IVF & Product Quantization
7. Choosing the Right Index
8. Batch Upserts & Data Pipelines
9. Namespaces, Sharding & Multi-Tenancy
10. Project — Multi-Source RAG Retrieval Service

By the end of this level you'll be fluent in four real vector databases, understand the index algorithms that make billion-vector search fast, and have built a retrieval service that pulls from multiple data sources.