Infrastructure Software Engineer, Dojo

About the position

As an Infrastructure Software Engineer on Tesla's Dojo team, you will build and optimize the systems that power our custom-built supercomputer. You'll focus on the core infrastructure and software layers that enable our machine learning workloads to run efficiently. You'll have the opportunity to contribute to Tesla's cutting-edge Dojo Supercomputer, working on low-level systems that will have a significant impact on autonomous driving, Optimus, and real-world AI applications.

Responsibilities

• Design and implement high-performance, reliable system software for our custom compute infrastructure
• Develop and optimize low-level drivers, runtime libraries, and tooling for the Dojo system
• Identify and resolve performance bottlenecks in distributed computing environments
• Build simulation and testing infrastructure to ensure reliability and performance
• Implement and improve monitoring, debugging, and profiling tools
• Work directly with hardware teams to optimize software-hardware interactions
• Collaborate with cross-functional teams to ensure that infrastructure meets the needs of compute-intensive workloads

Requirements

• Proven expertise in programming, ideally with hands on C++ development experience in a professional setting
• Demonstrated experience in one or more of the following: simulation software, compiler development or compiler toolchain engineering, systems programming and performance optimization, hardware driver development, build systems and infrastructure
• Experience with distributed systems and parallel programming
• Ability to work across software and hardware boundaries
• Strong problem-solving skills and attention to detail
• Degree in Computer Science, Computer Engineering, or equivalent hands-on experience