Member of Technical Staff, Infrastructure & Operations

Physical Superintelligence is a startup with roots at Google, NVIDIA, Harvard, Meta, MIT, Oxford, Johns Hopkins, Cambridge, and the Perimeter Institute building AI systems to discover new physics at scale. We are seeking engineers to build platform infrastructure at the intersection of computational science, AI systems, and software engineering.

Our mission is to discover and commercialize transformative physics breakthroughs at scale with artificial superintelligence, safely, verifiably, and for broad public benefit.

The last century's golden age of physics gave us transistors, lasers, and nuclear energy. We believe artificial superintelligence will unlock the next one. We're creating the infrastructure to industrialize scientific discovery and usher in this new era.

We have one product: new physics, at scale.

We are seeking a Member of Technical Staff, Infrastructure & Operations to own the path from commit to production for a codebase where most commits are written by AI agents, faster than any human reads them. Human review does not scale against that. CI does, and what CI enforces is the real architecture of how we ship: you decide what it enforces, and you build it, across an estate that spans GCP and specialist GPU providers, on a network you design and a failover you rehearse. You are building this from a thin starting point rather than inheriting a mature one. Every other team here ships through what you build.

• →

  Own the path from commit to production. Pipelines, staged environments, promotion with real gates. The artifact that passed the tests is the artifact that runs.

• →

  Run the cloud estate as code across GCP, AWS, and specialist GPU providers: org structure, IAM, quotas, and provisioning engineers use without filing a ticket. Repo and permission management counts as infrastructure here.

• →

  Design and operate the network underneath all of it: VPCs and subnets across providers, interconnects, DNS, cross-cloud failover. The failover gets rehearsed.

• →

  Make onboarding a system. How fast a new engineer becomes productive is an infrastructure property, not an HR one. You build the paved road: environments, docs-as-code, and the defaults that make the right thing the easy thing.

• Three or more years building and operating production infrastructure as code, with Kubernetes and at least one major cloud platform. You have shipped infrastructure other engineers depend on daily, and you have led incidents on systems you built.

• Three or more years of CI/CD ownership, including building it from zero to one rather than maintaining a mature system. Pipelines exist so engineers ship: the goal is safe speed, not fewer deploys.

• Depth in one of the following. We hire for one lane and expect you to learn the others on our platform:

  • CI/CD and release engineering: staged environments, promotion with real gates, build and registry economics.

  • Multi-account and multi-cloud networking: estates as code, VPC design, interconnects, DNS, and failover that gets rehearsed.

  • GPU cloud providers: the account, quota, and network realities of specialist providers running alongside a major cloud.

  • Observability and FinOps: OpenTelemetry, Prometheus, Grafana-class instrumentation, and cloud cost as a design input.

• A second lane from the list above.

• You have supported machine-generated or unusually high-velocity commit patterns.

• Paved-road work other engineers adopted unprompted: onboarding, environments, docs-as-code, defaults that make the right thing the easy thing.

• Toil reduced with code, and self-service abstractions preferred to tickets.

We hold a high technical bar and give people full ownership of their work, from spec to ship to on-call. We write contracts before logic, test against real systems instead of mocks, and favor simple designs that ship over clever ones that do not. Our development process is AI-native: we work with agentic coding tools daily, write specs that are legible to humans and agents alike, and lead with leverage.