You don't need an LLM for good emoji search: embeddings run faster and cost $0
maxleiter · x · 2026-09-27
Max Leiter pushes back on the hype around using Jev for classification tasks like Linear's emoji picker, arguing most such use cases can be matched or beaten with plain embeddings.
- He built semantic emoji search two years ago without any LLM: bge-small-en-v1.5 runs in-browser (37 MB, cached after first download), $0 per search
- His demo compares three approaches live with latency and cost: on-device bge-small, text-embedding-3-large + Postgres pgvector, and Jev via Vercel AI Gateway
- His rule of thumb: use Jev when you need actual LLM reasoning behind a decision; for semantic matching and retrieval, embeddings are faster, cheaper, and occasionally better
- He concedes embeddings require more setup (a database plus an embedding provider), while Jev is easier to wire up
More from coding & agent
- Adobe MAX: MCP across Creative Cloud, AI agents editing layers, Firefly Builder and Graph — umesh_ai · 2026-11-11
- Sydney Codex Community Buildathon set for Nov 14 with Cerebras-backed ultrafast compute track — paw_lean · 2026-11-11
- shadcn's improve hits 10k stars: let top models plan, cheap models execute code fixes — shadcn · 2026-11-11
- OpenAI open-sources Codex Security CLI and SDK for AI-driven vulnerability scanning and fixing — openai · 2026-11-11
- Claude Opus 5.5 runs an architecture search and finds AttnRes-Canon, a new top Transformer design — SonglinYang4 · 2026-11-11
- 5 research habits worth knowing: read papers backwards, reproduce first, wire coding agents via MCP — cneuralnetwork · 2026-11-11