About the Role

PolarGrid is building the infrastructure layer for real-time AI inference. We're looking for a Head of Infrastructure Partnerships to own our relationships with OEMs, bare metal providers, and AI marketplaces like OpenRouter.

This is a hybrid technical/commercial role. You'll be negotiating hardware specs and pricing one day, configuring inference marketplace listings the next. You'll be one of the first people shaping how PolarGrid sources compute and brings capacity to market.

What You'll Do

•⁠ ⁠Lead negotiations with OEMs and bare metal providers on pricing, SLAs, volume commitments, and specs
•⁠ ⁠Own our seller/provider presence on inference marketplaces
•⁠ ⁠Translate hardware capabilities into commercial offerings: pricing models, throughput guarantees, packaging
•⁠ ⁠Evaluate bare metal vs. colocation tradeoffs and drive procurement decisions
•⁠ ⁠Build and manage a pipeline of infrastructure partnerships that scale with our compute needs
•⁠ ⁠Work closely with engineering to make sure partner specs match what we can actually deliver

What We're Looking For

•⁠ ⁠Enough GPU/server hardware knowledge to negotiate intelligently (B300s, GB300s, networking, interconnects)
•⁠ ⁠Experience negotiating cloud or hardware contracts, OEM pricing, SLAs, volume deals
•⁠ ⁠Familiarity with the inference stack: vLLM, batching, throughput and latency tradeoffs
•⁠ ⁠Comfortable reading and redlining vendor agreements
•⁠ ⁠Strong grasp of inference unit economics: cost per token, utilization rates, margin
•⁠ ⁠Self-directed, you don't need a playbook, you'll write one

Bonus Points

•⁠ ⁠Existing relationships with bare metal providers
•⁠ ⁠Has set up a provider account on an AI marketplace before
•⁠ ⁠Startup or early-stage company experience

Why PolarGrid

We're a small team moving fast on a big problem. This isn't a supporting role inside a large partnerships org. You'll have real ownership over a critical part of the business.