🧮 Embeddings & Vector DBs
Embedding models, chunking, hybrid retrieval, vector store choice (Pinecone, Qdrant, Weaviate, pgvector), reranking.
The workflow
flowchart LR
A[Documents] --> B[Chunking<br/>strategy & overlap]
B --> C[Embedding model<br/>OpenAI / BGE / Cohere]
C --> D[Vector store<br/>Pinecone / Qdrant / pgvector]
D --> E[Hybrid search<br/>dense + sparse]
E --> F[Filter + rerank]
F --> G[Top-k results]
Embedding choice is rarely the bottleneck; chunking and reranking are.
Key takeaways
Videos (6)
Recsys Keynote: Improving Recommendation Systems & Search in the Age of LLMs - Eugene Yan, Amazon
Recsys is moving from hash-based IDs to learned multimodal semantic IDs that share representational space with LLMs.
Supabase Vector: The Postgres Vector database: Paul Copplestone
You probably don't need a dedicated vector DB — PG Vector inside the Postgres you already run gets you 90% of the way with RLS and joins.
The Hidden Life of Embeddings: Linus Lee
Treating embedding space as a steerable canvas with discoverable semantic directions unlocks UI patterns prompts alone can't deliver.
RAG and the MongoDB Document Model: Ben Flast
Co-locate embeddings and business data in MongoDB documents so RAG can blend transactional context with vector search via one query language.
The State of AI Powered Search and Retrieval — Frank Liu, MongoDB (prev Voyage AI)
Production AI search now means domain-specific embeddings, structured-filter hybrid retrieval, and agentic query decomposition — not one-shot vector search.
Vector Search Benchmark[eting] - Philipp Krenn, Elastic
Treat vendor vector-search benchmarks as marketing — only trust automated, reproducible, recall-anchored benchmarks across realistic read-write workloads.