NVIDIA is looking for a Senior AI Compute Engineer to join its Infrastructure Specialists team, where they will work on deploying and managing AI Compute infrastructure for various customers. The role involves interacting with customers and internal teams to implement large-scale AI Compute projects, requiring strong interpersonal skills and technical expertise in Linux environments.
Responsibilities:
- Primary responsibilities will include deploying, managing, and validating AI Compute/HPC infrastructure in Linux-based environments for new and existing customers
- Be the domain expert with customers during planning calls through implementation
- Handover-related documentation and perform knowledge transfers required to support customers as they begin rolling out some of the most sophisticated systems in the world!
- Provide feedback to internal teams such as opening bugs, documenting workarounds, and suggesting improvements
Requirements:
- 5+ years providing in-depth support and deployment services; solving problems for hardware and software products
- Knowledge and experience with Linux system administration, process management, package management, task scheduling, kernel management, boot procedures/troubleshooting, performance reporting/optimization/logging, network-routing/advanced networking (tuning and monitoring)
- Cluster management and provisioning technologies for bare-metal servers (bonus credit for BCM (Base Command Manager))
- Minimum of a four-year degree from an accredited university or college in Computer Science, Electrical or Computer Engineering or equivalent experience
- Scripting proficiency (Bash, Python, Ansible, etc.)
- Excellent interpersonal skills and the ability to deliver resolutions for customer issues as they arise
- Strong organizational skills and ability to prioritize/multi-task easily with limited supervision
- Experience with schedulers such as SLURM, LSF, UGE, etc
- An ability to travel to customer sites within the United States up to 30% of the time
- Experience with benchmarking tools such as HPL, NCCL tests, MLPerf as well as Kubernetes experience
- InfiniBand experience
- Experience with GPU (Graphics Processing Unit) focused hardware/software
- Experience with MPI (Message Passing Interface)
- Storage technologies such as Lustre or GPFS
- Familiarity with OEM GPU platforms