Engineering Manager, GPU Infrastructure
Why this team?
The GPU Clusters team builds and operates the superclusters that train Cohere’s frontier models. We sit at the intersection of hardware, distributed systems, and AI research. We work with cloud providers, researchers, and other infrastructure teams on problems few companies get to take on.
As an Engineering Manager, you’ll lead a team of engineers who care deeply about GPU infrastructure. You’ll set technical direction, grow people, and help the company scale a rapidly growing compute footprint.
As an Engineering Manager, you will:
- Hire, mentor, and grow a team of GPU infrastructure engineers, including performance, career development, and technical guidance on hard infrastructure problems
- Own the technical roadmap for the fleet: how we deploy, operate, and scale Kubernetes clusters, including workload scheduling, hardware fault detection, and performance
- Partner with researchers and ML engineers so the training and inference stack works well on new GPU architectures
- Work with cross-functional stakeholders such as Capacity, Finance, Legal, Security, and other infrastructure teams on planning, cost, compliance, and shared dependencies
- Drive operational excellence: observability for GPU utilization and reliability, automation of cluster provisioning, cost optimization, and vendor relationships
You may be a good fit if you have:
- Experience managing engineering or SRE teams, with a focus on technical mentorship, hiring, and growth, including in remote, distributed settings
- A background running large Kubernetes compute fleets in production, including in multi-cloud environments: multi-cluster operations, scheduling, node health at scale, and familiarity with IaC and infrastructure monitoring
- You’ve gone deep in one of the layers that make a GPU training fleet work, whether that’s cluster-wide operations, GPU networking, or hardware, and you’re willing to get hands-on and learn the rest
- Experience with cost optimization and capacity planning for GPU infrastructure
- A track record of partnering with researchers or ML engineers, and of making data-informed tradeoffs across reliability, cost, and delivery
COMPENSATION:
Cohere is committed to fair and transparent pay practices. The salary range listed for this role reflects the expected base compensation. Actual compensation offered will be determined by factors such as location, level, job-related knowledge, skills, education, and experience.
For candidates in the US, the Compensation Range is: $240,000 - $380,000 [USD]
For candidates in Canada, the Compensation Range is: $300,000 - $435,000 [CAD]