Build Local Image Search with Quarkus, ONNX, and pgvector (No Cloud Required)
Build a fast, offline image similarity search engine in Java using Quarkus, ONNX Runtime, LangChain4j, and PostgreSQL with pgvector. This hands-on tut...
Summary
The article explains how to build a semantic search engine using PartyKit, Cloudflare Vectorize, and Workers AI with about 160 lines of TypeScript.
It converts episode descriptions from the Braggoscope site into embedding vectors using the @cf/baai/bge-base-en-v1.5 model, stores them in a Vectorize index, and exposes a PartyKit server to handle indexing and nearest-neighbor queries. A small Remix-based web UI is used as an admin interface to manage indexing and test search behavior.