No harness optimization: agents just talked to each other via vanilla Codex and Claude Code
DimitrisPapail · x · 2026-09-15
Dimitris Papail adds technical details on his multi-agent math project: no harness optimization was done, using vanilla Codex and Claude Code. He simply asked the models to talk to each other with handoffs, but describes the process as very messy and not easily replicable—the result is reproducible, the process that produced it is not.
More from coding & agent
- Dev builds 3D universe explorer with 20 galaxies, 767 planets using Codex GPT-6.1 Sol — OpenAIDevs · 2026-11-02
- ChatGPT, Claude and Perplexity's fetchers don't run JavaScript, leaving client-rendered sites invisible to AI — gaganghotra_ · 2026-11-02
- SGLang boosts multimodal inference >10% by replacing GPU memory bookkeeping with a dict lookup — charles_irl · 2026-11-02
- kivi: hold Fn, talk, and it types polished text anywhere on your Mac — itsOmSarraf_ · 2026-11-02
- A base LLM plus 11 lines of code is all you need for a decision model — helloiamleonie · 2026-11-02
- Fix codebase structure before asking AI agents to write tests, says engineer — arpit_bhayani · 2026-11-02