Senior Systems Engineer, Amazon Special Projects
Core
Build the on-node runtime (agent daemon, sandbox, GPU manager, local storage) to enable reliable, low-latency ML inference at scale.
Role type
Senior Systems Engineer (Infrastructure)
Builds
Compute infrastructure for ML inference workloads
Domain
Cloud infrastructure, Machine Learning systems
Deliverable
infrastructure
Required skills
C++, Rust, Go, Linux internals, container runtimes, GPU management, systems design, automation
Preferred skills
Edge computing, fleet management, hardware attestation, storage performance profiling, daemon software development
Technologies
Linux, Docker, CUDA, NVML, nvidia-smi, TPM, TEE
Responsibilities
Design and implement agent daemon for GPU discovery and sandboxed execution; Architect local caching system with eviction policies and disk budget management; Build VRAM management and workload migration logic; Integrate hardware attestation and support fleet onboarding automation
