Optimization researcher: with LLMs and domain harnesses, we go 10x faster on harder problems
ErnestRyu · x · 2026-08-19
In a thread, optimization theorist Ernest Ryu described using LLMs for mathematical research, admitting some constructions were "frightening" and that he would not have figured them out without LLMs and a domain-specific harness. He argues that embracing these tools gives optimization theory a very optimistic future: going 10x faster, solving harder and more interesting problems, and having stronger downstream impact on optimization practice.
More from Models
- Prediction: frontier AI revenue growth could slow even as AI usage keeps rising — yi111 · 2026-10-16
- Nolla Health Post-Trains Nemotron 3.5 VL: Visual Diagnosis Jumps from 35% to 66.8% — LoganGrasby · 2026-10-16
- Dev argues coding models now beat nearly all engineers in every branch of software — zack_overflow · 2026-10-16
- Prediction: Chinese frontier models could match GPT-6 by March 2027 — teortaxesTex · 2026-10-16
- Liquid AI's d1-ultra, #1 on Artificial Analysis Decision Index, lands on Vercel — JosephJacks_ · 2026-10-16
- Weekly AI roundup: Claude Haiku 5.5 cuts prices ~90%, Gemini 4 Argon output jumps to 1M tokens — DeepLearningAI · 2026-10-16