Team open-sources Tahuna infra to power Hillclimb, an autonomous ML experimentation loop

Monaim101 · reddit · 2026-09-14

The team open-sourced Tahuna, their ML workload infrastructure, and started Hillclimb, an autonomous experimentation loop where an agent proposes and iteratively runs improvements.

Execution layer design:

Currently supports RunPod GPUs and R2 storage, with Docker self-hosting for the control plane; examples include SFT, RL agentic search, and MNIST. AGPL-3.0 licensed. The authors stress it's an early public preview and ask for feedback on inspectability and control of the execution layer.

Related event: Tahuna Open-Sourced: GPU Orchestration and Autonomous ML Experiments for Small Teams(2 posts)→

Original post →

More from coding & agent

coding & agent channel →