Training / AI Infrastructure

<h3>What You’ll Do</h3><ul style="min-height:1.5em"><li><p style="min-height:1.5em">Drive down wall-clock time to convergence by profiling and eliminating bottlenecks across the foundation model training stack stack, from data pipelines to GPU kernels</p></li><li><p style="min-height:1.5em">Design, build, and optimize distributed training systems (PyTorch) for multi-node GPU clusters, ensuring scalability, robustness, and high utilization</p></li><li><p style="min-height:1.5em">Implement efficient low-level code (CUDA, cuDNN, Triton, custom kernels) and integrate it seamlessly into high-level training frameworks</p></li><li><p style="min-height:1.5em">Optimize workloads for hardware efficiency: CPU/GPU compute balance, memory management, data throughput, and networking</p></li><li><p style="min-height:1.5em">Develop monitoring and debugging tools for large-scale runs, enabling rapid diagnosis of performance regressions and failures</p></li></ul><h3>What You’ll Bring</h3><ul style="min-height:1.5em"><li><p style="min-height:1.5em">Deep experience in distributed systems, ML infrastructure, or high-performance computing (8+ years)</p></li><li><p style="min-height:1.5em">Production-grade expertise in Python</p></li><li><p style="min-height:1.5em">Low-level performance mastery: CUDA/cuDNN/Triton, CPU–GPU interactions, data movement, and kernel optimization</p></li><li><p style="min-height:1.5em">Scaling at the frontier: experience with PyTorch and training jobs using data, context, pipeline, and model parallelism</p></li><li><p style="min-height:1.5em">System-level mindset with a track record of tuning hardware–software interactions for maximum utilization</p></li></ul><p>Find more <a href="https://www.arbeitnow.co.uk/english-speaking-jobs">English Speaking Jobs in United Kingdom</a> on Arbeitnow</a>