A graph-theoretic approach to building reliable LLM judges for retrievalgeorgianailab.substack.com·3 pts·_peregrine_·0
Mixing numeric attributes into text search for better first-stage relevanceturbopuffer.com·3 pts·_peregrine_·0
How to build a distributed queue in a single JSON file on object storageturbopuffer.com·2 pts·_peregrine_·0
Why BM25 queries with more terms can be faster (and other scaling surprises)turbopuffer.com·53 pts·_peregrine_·1
Vectorized MAXSCORE over WAND, especially for long LLM-generated queriesturbopuffer.com·2 pts·_peregrine_·0
Apache Kafka and Tinybird (ClickHouse) for Streaming Analytics HTTP APIskai-waehner.de·1 pts·_peregrine_·0