llnl / llnl/dmx-learn

Run benchmark review for orchestrator routing behavior

Open
#96 0 comments 0 reactions 0 assignees View on GitHub
agentic-dmx
Dominant language
Jupyter Notebook
Stars
10
Forks
1
Avg merge
1m
Merged PRs (30d)
34

Description

Run a focused manual review against the benchmark suite to validate intake,
lightweight EDA, and structure-first routing behavior in the orchestrator.

This issue is review-oriented. It should produce a narrow artifact describing
where routing still fails.

Scope:

- run benchmark prompts through the current skill set
- focus only on intake and model-routing behavior
- collect and categorize routing failures

Acceptance criteria:

- a review artifact exists
- routing failures are categorized clearly
- follow-up fixes can be opened as separate small issues if needed

---

Contributor guide

Open the contributing guide

Research direction

Start by locating the benchmark suite and current skill set, then run prompts focused on intake and model-routing behavior. Record the observed routing failures in a narrow review artifact, categorize them clearly, and ensure any follow-up fixes are described as separate issues.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
machine-learning
Issue type
Documentation
Difficulty
3/5
Estimated time
1-2 days
Activity status
Quiet
Clarity
Mostly clear
Newbie friendliness
62/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.