Senior Solutions Architect, AI Infrastructure

About the role

End-to-end cluster architecture is the core of this job: performance modeling, validation and first deployments of new GPU and networking designs. The person sits between engineering and the partners who build at scale.

At a glance

What you will do

• Work with cloud partners on GPU cluster and network design, sharing architecture guidance and sound build processes.
• Weigh design principles against real site limits to get clusters that perform well and stay supportable.
• Help partners debug cluster configuration and performance problems, using internal engineering knowledge and known bugs.
• Support early customer deployments of new GPU and networking architectures.
• Pass partner and field feedback on cluster design and workflows back to the engineering teams.

What they look for

• Degree in computer science, electrical or computer engineering, physics or a similar field, or equal experience.
• Eight or more years designing, validating and fixing GPU and HPC clusters.
• Record of designing large distributed systems, AI clusters or HPC infrastructure.
• Able to turn complex engineering ideas into clear documents, diagrams and reference material for customers.
• Can push partner issues to resolution together with product and engineering teams.
• Comfortable coordinating across customer, product, support and engineering groups.

Nice to have

• Experience leading bring-ups or builds of large AI or HPC clusters.
• Hands-on debugging of deployment issues on GPUs, NVLink and networking.
• Knowledge of NCCL, MPI, IMEX, NMX and collectives used in distributed training.
• Customer-facing background with good time management across several accounts.