YC F26's WonderSearch searches millions of documents without pre-embedding the corpus
KlausCodes · x · 2026-09-28
YC F26 startup WonderSearch launches query-time document search: no corpus-wide embedding step and no vector index to maintain, with claimed recall comparable to or better than embedding-based retrieval. Compute is spent only at search time, billed per search instead of upfront embedding costs — a notable challenge to the standard RAG pipeline for fast-changing or rarely-queried corpora. Launch credits included.
More from coding & agent
- anti-slop repo adds 23 LLM-judged oxlint rules to block AI-written TypeScript slop — teropa · 2026-11-16
- Open-source ComfyUI node enhances MiniMax H3 image conditioning for better texture and detail — Away_Exam_4586 · 2026-11-16
- Microsoft open-sources TokenEconomizer, cutting coding agent costs 18-31% — udmrzn · 2026-11-16
- Developer Builds a Shared Pixel World for Humans and AI Agents with Claude Code — BarcodeCutter · 2026-11-16
- 24 hours of prompting an AI agent yields GBC Factorio homage GBFactory — Dimillian · 2026-11-16
- Can AI auditors be talked out of their verdicts? An experiment in agent capture — gleech · 2026-11-16