DIRECT: retrieval-based next-question suggestions for conversational agents
burkov · x · 2026-10-16
Andriy Burkov outlines DIRECT, an end-to-end retrieval system for Next Question Suggestions (NQS). Users of shopping/support/healthcare agents often don't know what to ask, leading to dead-end conversations. Generating suggestions with an LLM at interaction time is expensive, slow, and risks unsafe questions, so DIRECT uses retrieval instead — but off-the-shelf embeddings miss domain language and the directional quality of a good follow-up, motivating a custom-built and evaluated system.
More from coding & agent
- Alibaba open-sources OpenGCUA, a full-stack generalist computer-use agent; +19.6 points on OSWorld-v2 across 8 models — Tongyi-MAI · 2026-10-16
- BotBeat: a social network where all users are AI agents, exposed via 32 MCP tools — Dismal_Piccolo4973 · 2026-10-16
- DKGI paper: agents write their own tools to search knowledge graphs, beating baselines by 6.7 Hit@1 — yuz9yuz · 2026-10-16
- Dev open-sources lgtmaybe: running PR code reviews on a local 27B model via hunk splitting — mattjcoles · 2026-10-16
- GLM 5.3 Flash is now free in OpenCode coding tool — DanielLockyer · 2026-10-16
- 12 React performance concepts in one infographic, from code splitting to Web Vitals — _jaydeepkarale · 2026-10-16