Senior HPC Operations Engineer

Lambda•California, PA

52d•Hybrid

About The Position

Lambda, The Superintelligence Cloud, builds Gigawatt-scale AI Factories for Training and Inference. Lambda’s mission is to make compute as ubiquitous as electricity and give every person access to artificial intelligence. One person, one GPU. If you'd like to build the world's best deep learning cloud, join us. Note: This position requires presence in our San Francisco/San Jose or Seattle office location 4 days per week; Lambda’s designated work from home day is currently Tuesday. Engineering at Lambda is responsible for building and scaling our cloud offering. Our scope includes the Lambda website, cloud APIs and systems as well as internal tooling for system deployment, management and maintenance.

Requirements

Are a deeply experienced HPC engineer comfortable with logical provisioning of a cluster
Have a strong understanding of HPC/AI architecture, operating systems, firmware, software, and networking
10+ years of experience in deploying and configuring HPC clusters for AI workloads
Have an innate attention to detail
Have experience with Bright Cluster Manager or similar cluster management tools
Are in expert in configuring and troubleshooting: SFP+ fiber, Infiniband (IB), and 100 GbE network fabrics Ethernet, switching, power infrastructure, GPU direct, RDMA, NCCL, Horovod environments Linux based compute nodes, firmware updates, driver installation SLURM, Kubernetes, or other job scheduling systems
Work well under deadlines and structured project plans also knowing when and how to ask for changes to project timelines
Have excellent problem solving and troubleshooting skills
Have flexibility to travel to our North American data centers as on-site needs arise or as part of training exercises
Are able to work independently and as part of a team
Are comfortable mentoring and supporting junior HPC engineers on cluster deployments

Nice To Haves

Experience with machine learning and deep learning frameworks (PyTorch, Tensorflow) and benchmarking tools (DeepSpeed, MLPerf)
Experience with containerization technologies ( Docker, Kubernetes)
Experience working with the technologies that underpin our cloud business ( GPU acceleration, virtualization, and cloud computing)
Keen situational awareness in customer situations, employing diplomacy and tact
Bachelors degree in EE, CS, Physics, Mathematics, or equivalent work experience

Responsibilities

Remotely deploy and configure large-scale HPC clusters for AI workloads (up to many thousands of nodes)
Remotely install and configure operating systems, firmware, software, and networking on HPC clusters both manually and using automation tools
Troubleshoot and resolve HPC cluster issues working closely with physical deployment teams on-site
Provide clear and detailed requirements back to other engineering teams on gaps and improvement areas, specifically in the areas of simplification, stability, and operational efficiency
Contribute to the creation of and maintenance of Standard Operating Procedures
Provide regular and well-communicated updates to project leads throughout each deployment
Mentor and assist less experienced team members
Stay up-to-date on the latest HPC/AI technologies and best practices