Achieving SOTA Deep Research: Training Models Over Scaffolding
SimonShaoleiDu · x · 2026-07-30
In an interview, Simon Shaolei Du shared how they achieved SOTA performance in deep research by training the model itself rather than relying solely on scaffolding. He noted that the system runs up to 150 sub-agents and treats verification as a distinct layer rather than a final prompt. He also referenced findings suggesting that a single agent, even with a one-million-token context, struggles to carry hard research problems.
Related event: Simon Du Reveals SOTA Deep Research via Model Training(2 posts)→
More from coding & agent
- jasonkneen Launches Browser Agent with 3D Voxel Avatars — jasonkneen · 2026-07-30
- Open Source Sticky Notes Plugin for Hermes Desktop Manages Fleeting Ideas — Teknium · 2026-07-30
- Agensis Launches Shared Workspace for Humans and AI Agents — jasonkneen · 2026-07-30
- Machinen Desktop: A Spatial Desktop Environment Built for AI Agents — kieranklaassen · 2026-07-30
- Dev Uses LLMs to Run Wayland and Linux Desktops Natively on iPad — maxleiter · 2026-07-30
- Developer Codes with AI for Two Days Using an Xbox Controller — noahsolomon · 2026-07-30