xAI Software Engineer at SpaceXSI to design and implement large-scale distributed systems powering supercomputing clusters. Focus on profiling, debugging, and optimizing performance across GPUs, Linux kernel, networking, and filesystems.
Responsibilities
Design, build, and implement a large-scale distributed system that powers one of the world's largest supercomputing clusters
Dive into the low-level stack to profile, debug, and optimize performance across diverse systems, including GPUs, Linux kernel, networking, and filesystems, to achieve peak efficiency
Collaborate on hardware, software, and algorithm co-design to push the boundaries of AI training
Maintain and innovate on our codebase to ensure scalability and reliability
Develop tools to enhance team productivity and streamline workflows
Qualification
Experience with Kubernetes (K8s)Collaborate in a fast-pacedProficiency in performance analysis
Required
5+ years of systems programming experience in C, C++, or Rust
Experience with Kubernetes (K8s), including cluster architecture, pod lifecycle, networking (CNI), storage (CSI), service mesh, and production-grade operations
Preferred
Computer systems fundamentals with a grasp of how computers execute code from transistors to high-level applications
Collaborate in a fast-paced, open environment to design and foundational systems
Strong debugging skills across the full stack — from kernel and OS up through container orchestration layers
Deep knowledge of operating systems internals (process scheduling, memory management, file systems, and synchronization primitives)
Proficiency in performance analysis, profiling, and low-level optimization techniques
Solid understanding of computer networks and the TCP/IP stack
Experience working with Linux kernel concepts or systems-level debugging tools (e.g., perf, gdb, strace, Wireshark)
Proficiency deploying and managing workloads using Kubernetes manifests, Helm, Operators, and GitOps workflows