triton , Linux kernel, Jax, Cuda, Tensorflow, Pytorch, Distributed Systems, Python, MPI, accelerator firmware, Libfabric, UCX, large-scale cluster software, GPU drivers, ROCm, vLLM, agile software development practices, GitHub Actions, oneAPI, SYCL, NCCL, RCCL

software engineering practices , Unix, Gcp, Linux, Pyspark, Spark, Azure, Python, AWS, CI CD pipeline, Model deployment

triton , Jax, Tensorflow, Distributed Systems, Cuda, Pytorch, Python, Linux kernel, ROCm, large-scale cluster software, GitHub Actions, vLLM, UCX, multi-node fabric-attached environments, MPI, accelerator firmware, RCCL, GPU drivers, multi-GPU, agile software development practices, NCCL, SYCL, Libfabric, oneAPI