Senior Performance Engineer, Efficiency Red Team
Core
Senior Performance Engineer leading deep technical research to discover and eliminate waste across the entire compute stack (CPU, DRAM, GPU, storage, I/O, networking) at Amazon's scale, turning findings into billions in capacity savings.
Role type
Senior hands-on IC performance engineer (efficiency/red team)
Builds
Fleet-wide optimization playbooks, automated profiling tools, and efficiency frameworks adopted across Amazon's infrastructure
Domain
Cloud infrastructure, hyper-scale compute, Generative AI, systems programming
Deliverable
production ML models | infrastructure
Required skills
Systems programming (C, C++, Rust, Java), performance profiling and analysis, root cause analysis across compute layers, design/leadership of systems architecture, mentorship
Preferred skills
GPU profiling (CUDA, memory management), Generative AI infrastructure internals, language runtime internals (JVM GC), capacity planning at hyper-scale, open source contributions
Technologies
C, C++, Rust, Java, CUDA, JVM, GPU, CPU, DRAM, storage subsystems, networking
Responsibilities
Apply first-principles analysis to uncover optimization opportunities; design rigorous investigations to establish root causes; build proof-of-concept implementations for novel efficiency approaches; develop repeatable patterns for fleet-wide optimization; quantify business impact in terms of capacity freed and dollars saved; collaborate across organizational boundaries; contribute to internal profiling tools and automated agents; influence performance engineering standards at scale