You will work with the latest Accelerated computing and Deep Learning software and hardware platforms, and with many scientific researchers, developers, and customers to craft improved workflows and develop new, leading differentiated solutions. You will interact with HPC, OS, GPU compute, and systems specialist to architect, develop and bring up large scale performance platforms. Does this sound like you? If so, we would love to hear from you!
What you will be doing:
Deploy, manage and maintain large scale HPC/AI clusters.
Managing Linux job/workload schedules and orchestration tools.
Support and maintain continuous integration and delivery pipelines.
Troubleshooting and fixing, bottom up from bare metal, operating system, software stack and application level.
Supporting Research & Development activities and engaging in POCs/POVs for future improvements.
What we need to see:
Bachelor's Degree in Computer Science, Engineering, or a related field; or equivalent experience.
5+ years of experience.
Knowledge of HPC and AI solution technologies from CPUs and GPUs to high speed interconnects and supporting software.
Experience with job scheduling workloads and orchestration tools such as Slurm, K8s.
Excellent knowledge of Windows and Linux (Redhat/CentOS and Ubuntu) networking (sockets, firewalls, iptables, wireshark, etc.) and internals, ACLs and OS level security protection and common protocols e.g. TCP, DHCP, DNS, etc.
Experience with multiple storage solutions such as Lustre, GPFS, zfs and xfs. Familiarity with newer and emerging storage technologies.
Python programming and bash scripting experience, automation and configuration management tools such as Jenkins, Ansible, Gitops.
Knowledge of Networking Protocols like InfiniBand, Ethernet.
Experience with virtual systems (for example VMware, Hyper-V, KVM).
Familiarity with cloud computing platforms (e.g. AWS, Azure, Google Cloud).
Ways to stand out from the crowd:
Knowledge of CPU and/or GPU architecture.
Knowledge of Kubernetes, container related microservice technologies.
Experience with GPU-focused hardware/software (DGX, Cuda).
Background with RDMA (InfiniBand or RoCE) fabrics.
