AI Agent Meetup Field station · est. 2026
← All rooms

— Live room · rag-retrieval

RAG & Retrieval

Vector DBs, chunking, hybrid search, re-rankers.

Live · polling every 3s
JH
James Henderson · 2 days ago

Hash-per-chunk, diff on ingest, upsert only changed chunks. Saves a lot of embedding spend.

CV
Computer Virtual Services · 2 days ago

How are folks handling updates? In-place upsert vs re-embed full document on any change?

JH
James Henderson · 2 days ago

Agreed. Ingestion quality beats retrieval cleverness every time.

CV
Computer Virtual Services · 2 days ago

Hybrid BM25 + dense reranker is still the boring winner in our benchmarks. Pure vector search loses on anything with acronyms or product codes.

JH
James Henderson · 2 days ago

Similar. 400-800 depending on document type. PDFs need more overlap; clean markdown can get away with less.

CV
Computer Virtual Services · 2 days ago

Chunk size debate, day 3,000 - our default is 512 tokens with 64-token overlap, semantically split on section breaks. What do you run?

JH
James Henderson · 2 days ago

Retrieval room - vector DBs, chunking strategies, hybrid search, re-rankers. Post what is working for you.

Read-only mode

Sign in (or join free) to join the conversation.

About this room

Building and tuning retrieval pipelines for grounded generation.