Bergum: Build RAG with BM25 first, hybrid search second, re-ranking third
Jo Kristian Bergum · The Rise and Fall of the Vector DB category: Jo Kristian Bergum (ex-Chief Scientist, Vespa) · Apr 19, 2025 · at 14:55
Former Vespa Chief Scientist Jo Kristian Bergum outlines the recommended architectural progression for developers building practical Retrieval-Augmented Generation (RAG) systems.
0:00 / 0:40exact quote · 40.8s
▶ Watch the full episode on YouTube → 720p mp4 · rendered on demand · StarZero watermark
“I think actually that a very strong baseline is the classical BM-Five like algorithm that's been around for 30 years, right? It's keyword matching, but it offers a very useful baseline for a lot of different search use cases because it gives you that baseline, right? Then you can start looking at using an off the shelf embedding model. To also embed the model and all of the engines, more or less, most of the engines have some kind of hybrid search capabilities, start to play with that. And then if you can afford it, both from a latency perspective and a cost perspective, you can look at adding like a re-ranking layer on top of that.”
quote is from the automated transcript, cleaned for reading: filler sounds and stutters are removed, nothing is rephrased. names can be misheard (the analysis reads context, assessments check outside sources). how →
More from Jo Kristian Bergum
Bergum: The standalone vector database infrastructure category is dying
“I'm not saying that the companies are dying, right? I'm just saying that the separate infrastructure category is dying, right? Because you have vector search capabilities in almost any DB technology nowadays, right?”
Bergum: Pinecone will not endure like MongoDB because it is too narrow
“So there's always like this convergence, but MongoDB kind of, it sticks, but I don't think that for Pinecoin that was originally leading that movement, it won't like stick in the same way.
It's too narrow.
It's too, too narrow.”
Bergum: pgvector is outpacing some standalone vector DBs in search capabilities
“So actually, what you can see, PJ Vector is doing more in the capabilities of vector search than some of the real vector database players, right?”
Bergum: In-database ML like PostgresML is the wrong architectural direction
“No, I'm not. I'm sorry. I, I'm not. I think this is, yeah, we also seen other players that tries to, you know, move a lot of the logic into the database, agentic embedding inference and whatnot. I think if the right direction is to Keep infrastructure a little…”
Bergum: Building knowledge graphs is the bottleneck in Graph RAG, not databases
“The core issue issue is actually to build the knowledge graph, right? The entities, the relationships. So if you say graph, graph, You know, databases or graph rag is going to kill vector rag and all that discussion. I think the first issue is to actually buil…”