HPC AI Cloud Engineer
- Hiring Organisation
- WWT EMEA UK LIMITED
- Location
- Manchester, UK
- Employment Type
- Full-time
inference, scientific workloads)Provision and optimize GPU/TPU-based infrastructure on GCP (A3/A4, TPU pods)Analyze performance across frameworks (PyTorch, TensorFlow, JAX, CUDA, ROCm)Identify system bottlenecks (compute, memory, network, I/O)Build automation tools for benchmarking and reportingCollaborate with teams to align workloads with optimal … architectureRequired SkillsStrong experience with GCP (Compute Engine, GKE, Storage, Networking)Hands-on with NVIDIA (CUDA/NCCL), AMD (ROCm), and TPUs (XLA/JAX/TF)Solid knowledge of HPC concepts (MPI, RDMA, InfiniBand, Slurm/Kubernetes)Experience with performance benchmarks (MLPerf, HPL, NCCL, STREAM)Proficiency in Python, Bash ...