Engineering Manager, GPU Infrastructure

Highlights

• Build and operate superclusters that train Cohere's frontier AI models.
• Sit at the intersection of hardware, distributed systems and AI research.
• Scale a rapidly growing compute footprint with researchers and cloud providers.

What they offer

• Equity

Ideal candidate

An infrastructure leader who loves mentoring distributed teams and going deep on hard systems problems. You care about operational excellence and partnering with researchers to scale frontier AI compute reliably and efficiently.

Requirements

• Experience managing engineering or SRE teams, with focus on mentorship, hiring and growth in remote, distributed settings
• Background running large Kubernetes compute fleets in production, including multi-cloud and multi-cluster operations
• Depth in one layer of GPU training fleets such as cluster operations, GPU networking or hardware, willing to learn
• Experience with cost optimization and capacity planning for GPU infrastructure
• Track record partnering with researchers or ML engineers and making data-informed reliability and cost tradeoffs