Overview In this role you build automation, diagnostics, and fleet-health infrastructure for AWS's AI/ML accelerator fleet. You own end-to-end systems spanning hardware to userland, crafting self-healing software that prevents failures at scale. You'll work across PCIe, GPU diagnostics, Linux drivers, and telemetry to isolate faults before they impact training jobs. This is a chance to influence architecture in a fast-moving, cross-functional environment and to shape industry-leading compute infrastructure.
Compensation / Benefits- sign-on payments and RSUs
- health insurance (medical, dental, vision, prescription, life/AD&D)
- 401(k) matching
- paid time off
- parential leave
- adoption and surrogacy reimbursement
Responsibilities- Build and own large-scale automation infrastructure for accelerator fleet health to enable zero-touch operations
- Design and implement test frameworks and diagnostic tooling for hardware validation and lifecycle coverage
- Develop predictive failure detection using telemetry, sensor data, and log correlation
- Create monitoring dashboards and real-time health alerts for manufacturing, lab, and production
- Define fleet health metrics (failure rates, MTTR, first-time fix rate, predictive accuracy)
- Debug complex system-level issues spanning hardware, firmware, kernel, drivers, and NICs
- Design scalable hardware bring-up tests and intelligent sequencing to shorten manufacturing cycles
- Develop data pipelines correlating test results and telemetry to identify systemic yield issues
- Maintain Linux device drivers for ARM/x86 and contribute to OS internals and GPU software stacks
- Support CI/CD pipelines and testing to deploy to manufacturing lines and production fleet
- Collaborate with ODMs and internal teams to ensure data/control path readiness and onboarding
- Contribute to post-incident reviews and create runbooks/documentation to sustain improvements
- Drive process improvements to increase agility and reduce development friction
- Occasional travel (
Key requirements- 6+ years of systems design, software development, operations, automation, and process improvement experience
- 6+ years of professional software development experience (non-internship)
- 6+ years of programming in languages such as C++, C#, Java, Python, Golang, PowerShell, or Ruby
- 6+ years of designing or architecting scalable and reliable systems
- Bachelor's degree in Computer Science, Computer Engineering, Electrical Engineering, or equivalent
- 5+ years of Linux/Unix operating systems experience
- excellent communicator
- ability to drive alignment across hardware, software, and operations teams
- cross-team collaboration
- Linux kernel/driver development
- PCIe/topology understanding
- GPU diagnostics and virtualization
