Arcee Team Shares Multi-Agent Coding Workflow for Complex Tasks
code_star · x · 2026-08-14
The author introduces an opinionated AI coding harness used by the Arcee team, designed specifically for long-running tasks. It can be used standalone or exposed as an abstraction layer to tools like Codex and Claude Code.
The author details a specific workflow: using a single Codex session to manage 4-5 parallel NAC sessions (running GLM 5.2 or Deepseek v4 Pro) to tackle complicated cross-repository problems. The harness is noted to shine when delegated large problems and left to operate autonomously.
Related event: Arcee Open Sources NAC Long-Running Agent Framework(14 posts)→
More from coding & agent
- DIRECT: retrieval-based next-question suggestions for conversational agents — burkov · 2026-10-16
- Claude Code and Codex now autocomplete the user's next action — tokenbender · 2026-10-16
- Hermes Agent ships official Reverse Engineer Anything plugin with one-command setup — Teknium · 2026-10-16
- MLExpert: 35B MoE Agent Trained to Autonomously Handle Machine Learning Engineering — Shangheng Du · 2026-10-16
- Cowork-Gym: Alibaba Synthesizes 64.9K Agent Work Environments, Boosting Qwen by 8.7 Points — Anni Zou · 2026-10-16
- MMSearch-OPD: Joint RL and On-Policy Distillation for Long-Horizon Multimodal Search Agents — PekingUniversity · 2026-10-16