Results for “pooling”

Optimizes Qdrant query performance for large limits across multiple shards by using Poisson-distributed subsampling to reduce inter-shard data transfer.

Provides persistent, compounding memory for AI agents across sessions using local markdown files, with optional Supabase-backed semantic search.