A tech-focused client in Singapore is seeking an MLOps Engineer to own and evolve their on-premise infrastructure for PyTorch-based training workloads. This role demands a strong focus on building maintainable training pipelines, optimizing compute workloads, and troubleshooting low-level systems issues. Ideal candidates should possess expert knowledge in PyTorch, advanced skills in Python and C++, and a solid understanding of networking and server tuning. Experience with job schedulers and cluster tools is preferred.
#J-18808-Ljbffr