Making AI search fast
I help teams build fast search over unstructured data: vector databases, RAG pipelines, and over a decade of web performance craft.
Fast typeahead search with Milvus
Milvus can serve typeahead search recommendations in about 4ms, but only one or two of the obvious approaches get the answer right. I tested 36 query shapes on Milvus 3.0 so you don't have to.
The RAG Cost Curve
When does a retrieval index actually beat live search? Measured break-even points, not vendor claims: 334 queries a day on a short agent history, never on code the model already knows, and 24 on code it has never seen.
More in the blog and the talk archive.