Amazon
Senior Systems Engineer, Amazon Special Projects
$151K–$205K
Seattle, WA · Senior · Full-time · Posted Oct 26, 2026
- Experience
- 6+ years in several areas, including programming, non-internship software development, system design, and systems design/software development/operations/automation/process improvement; 7+ years in systems programming.
About the job
Build and operate the on-node systems that enable reliable, secure, low-latency ML inference across compute hardware.
What you'll do
- Design and implement a Linux agent daemon for GPU discovery, node availability detection, and sandboxed inference workloads.
- Architect local model caching, including eviction, versioned manifests, disk budgets, and efficient updates.
- Build VRAM management and workload migration logic; integrate hardware attestation and support fleet onboarding.
What they ask for
- 7+ years of experience in systems programming (Rust, Go, or C++).
- Deep knowledge of Linux internals, including process scheduling, memory management, and storage subsystems.
- Hands-on experience with container runtimes and GPU tooling and management.
- Demonstrated ability to build reliable software for heterogeneous hardware and lead deployment of complex production software solutions.
Nice to have
- Experience designing caching systems with constrained storage budgets.
- Background in edge computing, fleet management, or embedded systems engineering.
Pay
151,200.00 - 204,600.00 USD annually (base salary)
Tools
Linux, C++, C#, Java, Python, Golang, Rust, CUDA runtime
AI focus
The systems support low-latency ML inference workloads; responsibilities include GPU discovery and management, runtime sandboxing, model asset caching, and VRAM management.
Model-written summary · the original posting is authoritative