Vast.ai
GPU Systems Engineer – HPC / Parallel Computing
A marketplace where anyone with spare GPUs rents them out by the hour and anyone who needs a GPU rents one, with the price set by supply and demand rather…
San FranciscoFull-time
Checked 22 Nov
What you’d do
- Design and tune GPU kernels and tensor libraries.
- Bring HPC techniques to AI inference at scale.
- Evaluate new chip designs and ways of managing resources.
- On site in San Francisco or Los Angeles.
What they look for
- Advanced C++ (17 or 20).
- CUDA, HIP, SYCL, OpenCL or similar.
- Strong systems optimisation and HPC performance tools.
Helpful: distributed training or inference frameworks.
Perks: health cover, 401(k) match, early-stage equity, meals on site.
