Show HN: Chisel – AMD GPU development that feels local but runs in the cloud

We've been running lots of AMD MI300X experiments lately. The price/performance is compelling compared to NVIDIA, and ROCm keeps getting better.

But, no local AMD hardware means constant SSH juggling. Upload code, compile remotely, run rocprof, download results, repeat. We were spending more time managing infrastructure than optimizing kernels.

Context: DigitalOcean's AMD GPU droplets ship with ROCm preinstalled. At $1.99/hour for MI300X access, it's cheaper than buying hardware. We're currently competing in the GPU mode kernel optimization competition. We spend a good chunk of our time setting up the infra to profile these kernels.

Solution: Built Chisel to make AMD GPU development feel local. One command spins up a droplet, syncs your code, runs profiling with rocprof, and pulls results back. It handles the SSH, rsync, and teardown automatically.

  Key features:
  - chisel up creates MI300X droplet in seconds
  - chisel sync pushes only changed files
  - chisel profile kernel.cpp compiles, profiles, downloads traces
  - chisel pull grabs artifacts back to local
  - Auto cleanup prevents zombie droplets

The profiling integration was the killer feature for us.

Available on PyPI: pip install chisel-cli

Source: https://github.com/anthropics/chisel

Would love feedback, especially from people doing GPU work and exploring AMD alternatives to NVIDIA.

If you're interested in contributing, we'd welcome any help. our general short-term direction is to add grafana, concurrent runs, better error handling, and support for other cloud providers.