Inference Software Engineer - Collectives at Etched.ai

Candidates should possess strong proficiency in Rust and/or C++, familiarity with PyTorch and/or JAX, and experience designing/optimizing collectives such as NCCL, MPI collectives, and XLA collectives. Solid systems knowledge, including Linux internals, accelerator architectures (e.g., GPUs, TPUs), high-speed interconnects (e.g., NVLink, InfiniBand) and RDMAS are required, along with a strong understanding of distributed systems concepts, algorithms, and challenges. Experience analyzing performance traces and logs from distributed systems and ML workloads is also necessary.