redhat-et / redhat-et/code-agent
Minimal loop for RL training
Open
@TreeinRandomForest is already working on this.
Since Apr 6, 2026.
- Dominant language
- Python
- Stars
- 1
- Forks
- 4
- PR merge metrics
- No merged PRs in 30d
Description
See branch "minimal" (in progress) for a simple and direct RL loop that can be run on a bare-metal machine:
Loop:
Prompt -> LLM -> Patch -> Verifier Hierarchy -> Feedback appended to prompt
End in K iterations
Requirements:
- Verifiers are self-contained i.e. can be containerized easily
- Focus on SWE-bench
- Focus on OpenRLHF
Contributor guide
No contributing guide indexed for this repository
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Assessment
This issue has not been assessed yet.