Skip to content
For ML engineers

Can you reproduce what your
agent just did?

Run this once. See everything that actually happened.

$uv tool install roar-cli
or: pipx install roar-cli  ·  pip install roar-cli

No account required·~30 seconds·works on your existing code

Run this on your last training job.
roar in action
~/mnist
What roar captures
[ DAG image — https://treqs.ai/hubfs/raw_assets/public/example_dag.png ]

End-to-end lineage for every run — so you can rebuild any artifact, any time. GLaaS →

"We can't reproduce a model from three months ago. Would be great to have the recipe & datasets stashed."

— CTO, Series B robotics company  →
No changes to your workflow
No code changes
Works with PyTorch, JAX, TensorFlow
No YAML
Works with Ray / multi-node
No account required
<3% overhead · benchmarks →

ready to see it?

"I'm downloading it today."

— ML Engineer, robotics startup
in your repo
$uv tool install roar-cli
$roar init
$git commit -a -m "Add .roar to .gitignore"
$roar run python train.py
or: pipx install roar-cli  ·  pip install roar-cli

No account required·~30 seconds·works on your existing code

Run this on your last training job.