xAI · Foundation models
Software Engineer, Linux Kernel
- Seattle, +1 more: Palo Alto
AI roleMLOpsML platform
Still open on xAI's site · checked just now
- Posted
What you'll do
As written by xAI · Read the original
SOFTWARE ENGINEER, LINUX KERNEL
At SpaceXSI, we design, build, and operate the largest AI supercomputers in the world, from the ground up. As a member of the Platform Core team, you will be responsible for the lowest layers of the software stack on the devices setting the pace for how fast we can push the frontier of AI. Our team’s software interfaces directly with the core hardware of the supercomputer and its network fabric. Our mission is to expose the full capability of that hardware — limited only by physics — to the rest of the stack, all while retaining the smallest, most robust footprint possible.
RESPONSIBILITIES:
- Build and maintain the lean, high-reliability Linux-based operating system that underpins our supercomputer network fabric
- Write (or rewrite) high-performance device drivers to extract the maximum physical capability from our network and compute hardware
- Manage deployment, operation, and debugging of our platform in production, including: rigorous resolution and root-causing of anomalies to ensure the system is constantly improving in reliability and developing and constantly improving our tools for observability and configuration management
- Collaborate with our hardware teams and external partners to design the next generation of supercomputer hardware – and then bring up our OS and software stack on it
Requirements
BASIC QUALIFICATIONS
- Hands-on systems programming experience in C or C++
PREFERRED SKILLS AND EXPERIENCE
- Strong operating systems fundamentals (scheduling, memory management, I/O) — you must be able to articulate in extreme detail how a computer works, from the hardware all the way up to high-level applications
- Understanding of computer networking, both in the host (OS stack) as well as common network devices like switches and routers
- Excellent written and verbal communication skills
- Deep knowledge of the Linux kernel and its networking stack — transferable experience with another production-grade kernel is okay, too!
- Proficiency with debugging tools spanning kernel, network, and userspace (e.g., ftrace, perf, Wireshark/tcpdump, eBPF, gdb, and more)
- Experience bringing up new hardware from scratch — you have flashed bare-metal software on a microcontroller with JTAG, or bootstrapped a Linux kernel and rootfs on new hardware
- Demonstrated track record of developing and optimizing device/peripheral drivers — ideally for high-speed interfaces, or with advanced considerations like DMA and cache coherency
- Understanding of major peripheral hardware interface standards (Ethernet, PCIe, I2C, SPI, etc.) — you can critically review a schematic, and debug these interfaces if needed
- Experience with product security, OS/system hardening, and secure boot schemes
Skills
Stack
- C++
- C
Working languages
English
Salary & benefits
COMPENSATION AND BENEFITS
$180,000 - $440,000 USD
Base salary is just one part of our total rewards package at SpaceXSI, which also includes equity, comprehensive medical, vision, and dental coverage, access to a 401(k) retirement plan, short & long-term disability insurance, life insurance, and various other discounts and perks.
SpaceXSI is an equal opportunity employer. For details on data processing, view our Recruitment Privacy Notice.
Company
Last checked . The details above were extracted automatically; the employer's original posting is the authority.