Description
Achira is hiring a GPU Optimization Engineer to engineer high-efficiency implementations of advanced deep-learning architectures for foundation simulation models in molecular machine learning. The role involves profiling and optimizing PyTorch and JAX, accelerating performance-critical code with CUDA, Triton, Warp, and related frameworks, collaborating with NVIDIA, and working with scientists to identify high-impact optimization opportunities. Candidates need at least two years of professional GPU optimization experience, strong GPU programming fundamentals, observable optimization artifacts, and experience collaborating on multi-person software projects. The role is based in San Francisco or New York City, with full-time remote candidates considered if they are willing to travel frequently.
