Systems Engineer Intern
About the role
Dedalus Labs builds persistent computers for AI agents. The internship focuses on improving the control plane or Rust VMM, with hands-on bug fixing, testing, and performance work under guidance from engineers. The role emphasizes learning the codebase, tackling difficult problems, and delivering a working, well-documented result.
Responsibilities
- Fix bugs and papercuts, simplify code, improve tests and tooling, and make focused speed improvements.
- Work within existing systems with support from an engineer.
- Conduct reproducible investigations into bottlenecks (e.g., slow startup, restore, I/O paths) and measure focused changes.
- Implement runtime or storage improvements with tests for cancellation, resource cleanup, or recovery after a process stops.
- Develop failure-injection tests to expose incorrect ordering or lifecycle assumptions and verify repairs.
- Perform small refactors to remove repeated work or reduce allocations and improve measurable slow code paths.
- Deliver a completed scope with code, tests, explanations, and handoff for others to continue.
Qualifications
- You can build and debug software in Rust, Go, C, C++, or similar languages. Rust preferred but not required.
- Working understanding of memory, concurrency, data structures, and OS behavior relevant to the project.
- Experience with a small storage engine, networking tool, runtime, compiler, OS component, or performance investigation is a plus.
- Ability to explain what you implemented and what you learned beyond initial instructions.
- Ability to narrow problems, read source/docs, run experiments, and learn from feedback.
- Willingness to use AI coding tools when helpful and to explain, test, and correct code.
Skills
Benefits
- Paid internship.
- In-person in San Francisco.
- Flexible start dates and term length (3-9 months).
- Opportunity for full-time conversion; no guaranteed conversion.
About Dedalus Labs
Connect any LLM to any MCP tools and create complex agents in minutes.
Dedalus Labs is building Vercel for AI Agents. We host MCP servers on our cloud and handle the boring stuff like autoscaling and load balancing so users can go from idea to production in a single click. Our OpenAI-compatible SDK lets users orchestrate complex agentic workflows through one clean API. Route between any LLM provider, connect any MCP tools from our hosted marketplace, and ship in minutes.
More open roles at Dedalus Labs
- Software Engineer, Infrastructure & Performance3w agoSan Francisco
- Systems Engineer1mo agoHybrid (San Francisco) · Engineering
- Infrastructure Engineer Intern1mo agoHybrid (San Francisco) · Engineering