Sr. Software Engineer, Network (C++)

xAI•Seattle, WA

About The Position

At SpaceXSI, the Colossus team designs, builds, and operates the full AI supercomputer stack, including massive GPU clusters, high-speed interconnect fabric, and the software that supports them. Colossus powers Grok and other frontier AI models using a custom high-performance datacenter network designed to provide ultra-low latency and high bandwidth across hundreds of thousands of GPUs. As a Software Engineer on the Colossus Networking team, you will develop core networking software to maximize the performance and reliability of the datacenter fabric. The work affects training efficiency, model convergence, and AI development speed. Engineers own the full software lifecycle, from design and implementation through deployment, monitoring, and iteration based on real-world performance at scale. The role involves distributed systems, high-performance networking, and real-time control of a large AI supercomputer.

Requirements

Bachelor’s degree in computer science, engineering, math, or another technical discipline; alternatively, 5+ years of professional software development experience in lieu of a degree.
Development experience in C or C++.

Nice To Haves

Strong professional experience writing high-performance C/C++ in production environments.
Experience developing, debugging, and deploying software that runs at scale in real-world systems.
Deep knowledge of networking protocols such as UDP, TCP/IP, and RDMA, distributed systems, and large-scale datacenter fabrics.
Background in real-time systems, high-performance computing, low-latency networking, or resource-constrained environments.
Creative problem-solving ability, exceptional analytical skills, and strong engineering fundamentals.
Excellent written and verbal communication skills.
Ability to thrive in a fast-paced, dynamic environment with evolving requirements.
Experience with security considerations in large-scale distributed systems.

Responsibilities

Develop routing and traffic-engineering algorithms for the Colossus high-performance datacenter network.
Develop highly reliable, real-time software for switches that form the backbone of the low-latency, high-bandwidth AI training fabric.
Participate in and lead architecture, design, and code reviews.
Develop prototypes and run experiments to validate design decisions at small and full-cluster scale.
Build tools for software development, deployment, data analysis, visualization, and testing across virtualized environments, hardware-in-the-loop setups, and live production clusters.
Deploy reliable software updates through continuous integration and release systems with rigorous testing and monitoring.