1 to 4 of 4 Permanent Slurm Workload Manager Jobs in London

systems, CI/CD, etc.) Attention to detail needed to manage and debug production services. Experience with research clusters and implementing tools such as Slurm workload manager. Job Duties Own the lifecycle of our Linux-based servers and applications across our multiple business environments. Automate and troubleshoot a more »

magnitude of training runs Explore novel synthetic data generation techniques Engineer robust, high-performance inference Experience Technical: Have experience operating orchestration systems such as SLURM, Ray, or similar. Experience in creating and managing multi-instance clusters for data and model parallel training across GPUs/TPUs, preferably using PyTorch more »

of your team, ideally for a l eading AI research laboratory, or a pioneering AI business Key Requirements: Python and PyTorch expertise Experience in SLURM, Ray, or similar Graphics Processing Units (GPUs) Experience in creating and managing HPC clusters for ML models Experience in efficiently serving large ML models more »

high-performance inference platforms Collaborate in defining and steering their evolving inference and training stack Experience Technical: Have experience operating orchestration systems such as SLURM, Ray, or similar. Experience in creating and managing multi-instance clusters for data and model parallel training across GPUs/TPUs, preferably using PyTorch more »