Software Engineer, Infrastructure & Performance

Dedalus Labs builds persistent computers for AI agents. Our flagship product, Dedalus Machines, gives agents an isolated environment where they can run software, keep files and state, and work over time. We're looking for an infrastructure engineer focused on the team's development speed, including the systems that let engineers build, test, deploy, and operate their work. This is a full-time, in-person role in San Francisco.

Responsibilities

Make the development loop faster by profiling builds, tests, CI queues, artifact transfers, and local workflows; use and improve Bessemer (BSMR) and Hollywood; design APIs, controllers, CLIs, and services that support multiple internal products and consumers; improve CI/CD, GitOps, environment provisioning, staging, production rollout, and recovery; build and operate fast infrastructure including Linux servers, containers, virtual machines, and Kubernetes environments; work with the in-office homelab to provision, maintain, and test machines and networks

Requirements

Strong software engineering skills; experience building or operating infrastructure that other people rely on; depth in Go, Rust, C++, or another systems language; practical Linux debugging; thoughtful API design; ability to turn an incomplete goal into a concrete problem; ability to measure performance changes and verify results; ability to explain and support past work with evidence

Nice to have

Experience with Buck2, Bazel, Blaze, remote execution, or build cache design; experience scaling production services to 100,000 or more requests per second; hands-on hardware and networking experience; experience with Ethernet, SFP+/QSFP optics, direct-attach cables, RDMA, or RoCE; familiarity with Terraform; experience with build systems, CI/CD, GitOps, Kubernetes, observability, and infrastructure as code; TypeScript experience

Benefits

Relocation assistance; visa sponsorship