Machine Learning Systems Engineer

• 1+ years in deep learning training or inference systems, or distributed systems
• Hands-on experience with CUDA, Triton, PyTorch, NCCL or RDMA
• Someone who gets genuinely excited about shaving milliseconds off production workloads

Would you rather make a video model 10x faster than train one?

We’re partnering with a seed-funded AI infrastructure startup that’s rebuilding the training and inference stack for world models, from the GPU kernels all the way up to distributed serving. Today’s ML stack was built for language models. This team is rebuilding it for image, video and world-model workloads, and their public benchmarks already show order-of-magnitude speedups at a fraction of the cost.

The team is five cofounders with deep expertise across distributed systems, kernel optimization, cloud infrastructure and research. You’d be one of the first engineering hires, reporting directly to the CEO.

What you’ll do:

• Optimize GPU and system performance for image, video and world-model training and inference
• Profile and remove bottlenecks at the kernel, memory, system and cluster level using Nsight and related tooling
• Write low-level CUDA and Triton optimizations on code paths that run in production
• Build distributed inference and training engines for diffusion models across multiple GPUs and nodes
• Own communication performance: NCCL, RDMA over InfiniBand/RoCE, and disaggregated serving
• Build benchmarking and regression harnesses so performance gains stick