OpenForge RL brings end-to-end training to harness-based AI agents
burkov · x · 2026-07-26
Modern AI agents increasingly depend on elaborate inference harnesses such as Claude Code or Codex to handle multi-turn reasoning, tool use, and access to external systems.
The paper argues that these stateful, multi-process harnesses are hard to train end-to-end with today’s open SFT/RL stacks, because they cannot naturally express that kind of inference setup. To address that gap, researchers from Columbia and Microsoft present OpenForge RL, an open-source framework for training harness-based agents end-to-end across diverse environments.
Related event: Microsoft Open-Sources OpenForgeRL for End-to-End Agent Training(4 posts)→
More from coding & agent
- Someone built a full 3D Doom-style game with SQL queries — thursdai_pod · 2026-07-26
- Google ADK talk shows how managed MCP servers connect agents to real-world data — kuanhoong · 2026-07-26
- Google ADK hands-on lab teaches 55-plus attendees how to test and guard AI agents — kuanhoong · 2026-07-26
- Agent sandboxes are everywhere, but authorization is the harder problem — ashsg2016 · 2026-07-26
- “The classic /loop” turns the agent-environment feedback cycle into a meme — dee_hw · 2026-07-26
- Claude warns before it hits context limits, unlike Codex’s more seamless flow — steipete · 2026-07-26