Software Engineer, Linux Kernel (Starlink)
SpaceX (Platform Core / Starlink) seeks a Linux kernel engineer to build the low-level OS powering its AI supercomputer network fabric and the Starlink satellite laser mesh. Day to day: writing high-performance device drivers in C/C++, debugging production systems, and bringing up new hardware alongside hardware teams.
- Compensation
- $125K–$200K / yr
- Location
- Bellevue, WA
- Work setup
- Not specified
- Seniority
- Intern
- Employment
- Internship
- Experience
- 2+ years
- Category
- Software engineering
- Posted
Skills
Must have
- Bachelor's degree in computer science, engineering, math, or science discipline; OR 2+ years of professional software development experience in lieu of a degree (internship experience applicable)
- Hands-on system programming experience in C or C++
- Must meet ITAR requirements: U.S. citizen/national, lawful permanent resident, refugee, or asylee (or eligible for required authorizations)
Nice to have
- Deep knowledge of the Linux kernel and its networking stack, or experience with another production-grade kernel
- Proficiency with debugging tools spanning kernel, network, and userspace (ftrace, perf, Wireshark/tcpdump, eBPF, gdb)
- Experience bringing up new hardware from scratch (flashing bare-metal software with JTAG, bootstrapping a Linux kernel and rootfs)
- Track record developing and optimizing device/peripheral drivers, ideally for high-speed interfaces or with DMA and cache coherency
- Understanding of peripheral hardware interface standards (Ethernet, PCIe, I2C, SPI); can critically review a schematic and debug interfaces
- Experience with product security, OS/system hardening, and secure boot schemes
Full description
SpaceX was founded under the belief that a future where humanity is out exploring the stars is fundamentally more exciting than one where we are not. Today SpaceX is actively developing the technologies to make this possible, with the ultimate goal of enabling human life on Mars.
SOFTWARE ENGINEER, LINUX KERNEL (STARLINK)
SpaceX was founded under the belief that a future where humanity is out exploring the stars is fundamentally more exciting than one where we are not. Today SpaceX is actively developing the technologies to make this possible, with the ultimate goal of enabling human life on Mars.
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. **
RESPONSIBILITIES:**
- Build and maintain the lean, high-reliability Linux-based operating system that underpins our supercomputer network fabric and Starlink laser mesh of satellites.
- 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.
- 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.
BASIC QUALIFICATIONS:
- Bachelor’s degree in computer science, engineering, math, or science discipline; OR 2+ years of professional experience in software development in lieu of a degree (internship experience is applicable).
- Hands-on system programming experience in C or C++.
PREFERRED SKILLS AND EXPERIENCE:
- Deep knowledge of the Linux kernel and its networking stack or experience with another production-grade kernel.
- 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.