Skip to content

Introduce CodeClash Ladder#94

Open
john-b-yang wants to merge 5 commits into
mainfrom
john/ladder
Open

Introduce CodeClash Ladder#94
john-b-yang wants to merge 5 commits into
mainfrom
john/ladder

Conversation

@john-b-yang

Copy link
Copy Markdown
Contributor

This PR introduces the assets for creating and running models against CodeClash ladders.

To learn more about "CC:Ladder", check out the blog post (forthcoming): https://codeclash.ai/insights/20260115_human_ai_ladder/

Specifically, we introduce:

  • Ladders for Core War (262 solutions), RobotRumble (58 solutions)
  • run_ladder.py script for running model against ladder
  • Minor tweaks to existing player.py file to make branch checkouts smoother
  • Minor tweaks to robotrumble.py to make validation more general

@john-b-yang john-b-yang requested a review from klieret January 7, 2026 05:08

@Muhtasham Muhtasham left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Reviewed against current upstream/main.

Verification:

  • uv run pytest -q tests/arenas/test_robotrumble.py
  • uv run ruff check codeclash/agents/player.py codeclash/arenas/robotrumble/robotrumble.py scripts/run_ladder.py
  • parsed ladder configs through resolve_includes
  • uv run python scripts/run_ladder.py --help
  • uv run pytest -q

Full suite passed: 173 passed. I did not find a blocking issue. Non-blocking follow-up: configs/ablations/ladder/README.md still describes the old all-pairs runner flow, while this PR turns scripts/run_ladder.py into a ladder-climb runner.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants