Lead Engineer, AI Infrastructure (C++, C)

SpaceXSI•Palo Alto, CA

14h•$220,000 - $357,000

About The Position

SpaceXSI is seeking a lead software engineer for its AI infrastructure team supporting the Grok training program. The role leads the team that architects, implements, and maintains software infrastructure for training Grok models across large compute clusters, potentially involving hundreds of thousands or millions of GPUs. The software is expected to deliver reliability, fault tolerance, throughput, security, and a strong developer experience. The lead provides technical direction, removes engineering hurdles, and works with teams across model engineering, kernels, networking, hardware design, platform engineering, and site operations.

Requirements

Bachelor’s degree in computer science, engineering, mathematics, or another technical discipline.
Hands-on systems programming experience in C or C++.

Nice To Haves

Developed, debugged, and deployed software used in real-world applications or projects.
At least 3 years of software engineering experience, or at least 1 year of C/C++ experience at SpaceX.
Solid fundamental knowledge of NVL72 GB300 racks, computer architecture, networking, mathematics, and physics.
Strong skills in modeling, debugging, performance optimization, and unit testing.
Creative problem-solving approach, exceptional analytical skills, and strong engineering fundamentals.

Responsibilities

Develop reliable, performant, and maintainable software for training Grok models in SpaceX's compute clusters.
Provide technical direction for future training architectures and hardware designs.
Work closely with model engineers, kernel engineers, network engineers, hardware design engineers, platform engineers, and site operations engineers.
Develop prototypes to validate key design concepts and quantify technical constraints.
Lead team planning with stakeholders in other teams and management, develop milestones, and execute against them.