Turbopuffer cofounder confirms Cursor dropped server-side RAG for grep as models improved
ivan_bezdomny · x · 2026-09-29
Turbopuffer cofounder Simon (@Sirupsen) confirmed that Cursor called in July to say their new models' code evals no longer showed the same gains from semantic search as previous models.
Key points:
- Selecting the right context in a single turn used to be huge, but modern code-local agents can answer complex questions with simple tools like grep(1).
- In local codebases with small data — and where you need a full local copy anyway to edit and run — offloading to search infrastructure is overkill. Grep works fine, as it has for expert programmers for decades.
- Search engines remain necessary only for large-scale data.
Context: Turbopuffer's S3-based infra had been serving Cursor's server-side RAG search. The shift is a notable signal for both RAG infrastructure vendors and coding-agent architecture.
More from coding & agent
- Swargs claims $50M+ transaction volume and 6,000+ listings on its AI agent marketplace — KyeGomezB · 2026-09-29
- Sonnet 5.5 effort settings make no difference in 15-task coding test: 9/15 at low, medium and high — every · 2026-09-29
- Claude Connectors: 7 essential integrations to stop copy-pasting between Claude and your apps — Aiden_Tech_Ai · 2026-09-29
- AI made building so cheap that 5 launches beat 5 months of thinking — alexmacgregor__ · 2026-09-29
- AI agent buys NBA tickets end-to-end, pivots on its own after seats sell out mid-checkout — armand_ruiz · 2026-09-29
- Embedded dev on Chinese models: DeepSeek and GLM are good enough, skip Codex/Claude — sven_ai · 2026-09-29