Senior Infrastructure Engineer

The Role

We train action policies and world models on the largest corpus of action-labeled gaming data in the world, and we run Medal, a platform used by millions of gamers every day. Both depend on the same infrastructure: billions of gameplay clips a year that have to be stored, moved and served, under research workloads one day and consumer traffic the next.

We're looking for a generalist infra engineer to own that end to end, with cloud and storage at the center: the cloud architecture and the fleet it runs on, object storage for petabytes of gameplay video, and the data pipelines and CI/CD around them. You'll make the architecture calls before scale turns them into emergencies, plan for where today's systems break at 10x the data, traffic or team, and go from infra design straight into application code when that's what the problem needs.

You'll also build our infrastructure function in New York: set the technical direction, hire and grow the team, and define how we design, review, deploy and run systems, while staying hands-on yourself. You'll work directly with research and product engineering across General Intuition and Medal, and be the person they pull in for system design.

What We're Looking for

• Deep, hands-on cloud and storage experience, run in production. We run on GCP and CoreWeave; what matters is that you've owned Terraform, IAM and Kubernetes on a major cloud, and object storage at a scale where placement, tiering and egress cost were your decisions.
• 6+ years of hands-on exposure to planet-scale distributed systems (search infra, large scale storage or query systems, time series systems, or similar), with strong fundamentals in storage, query engines, consistency, and running these systems in production.
• A track record of building infra from scratch, not just operating what someone else designed, ideally as a founder or one of the first infra engineers, and evidence that you chose to keep building hard things.
• Experience leading infra while staying hands-on: you've set technical direction and raised the bar of the engineers around you, ideally including hiring them, and you still write code today.
• Range across the whole stack, from low-level infra up through application code. You shouldn't need a handoff to build the thing your infra supports.
• High agency and good judgment under ambiguity. Priorities shift between a research lab and a live consumer product, and you work out what actually matters and do it.

Nice to Have

• Video or media at petabyte scale, or genuine games practice and familiarity with gaming data.
• Adjacent depth in ML tooling (PyTorch or JAX, world models), data platforms / orchestration (e.g., Dagster), GPU training clusters, or systems work in Rust or C++, including serious open source contributions.
• Experience running infra that serves both research workloads and consumer-facing production traffic at the same time.