DuoDiscover Tops NVIDIA's SOL-ExecBench, Speeds TPU Kernels 2.68x by Making Evaluation Part of Discovery
adityagp · x · 2026-10-15
Dilara Soylu introduces DuoDiscover, an AI agent system that tackles engineering's core bottleneck: agents can generate ideas in seconds, but evaluating them takes hours or days. Its trick is making evaluation part of what agents discover.
Highlights:
- #1 on NVIDIA's SOL-ExecBench
- 2x the experts on SWE-fficiency
- 1.92x speedup on vLLM; 1.44x on FreeToken
- 2.68x over production TPU kernels
- Beats the best human- and AI-designed CPU prefetchers
The core idea: instead of generate-then-test, let the cost of evaluation itself become an object of the agent's search.
More from coding & agent
- Dev's honest scoreboard: pattern rules catch under 25% of prompt injections, guard model catches most — colinmcnamara · 2026-10-15
- SpaceX's Cursor acquisition hailed as one of the fastest, best-executed integrations — kieranklaassen · 2026-10-15
- cantrell on Gemini 4 Argon: heavy thinker, building long decode to push past 64K token limit — cantrell · 2026-10-15
- Creator Shares Fix for AI Video Voice Consistency, Compares Claude vs ChatGPT vs Figma Make — Tegadesigns · 2026-10-15
- Indie dev builds AI game in 4 days: 297 Magnific assets, one-prompt Meshy animations — techhalla · 2026-10-15
- Google's AIS Team Hiring FTEs and Interns for AI-Infra Agents and Verifiable Reasoning — VioletNPeng · 2026-10-15