Senior Embedded Performance Engineer
Core
Write and optimize performance-critical ML kernels and low-level control paths for AI inference hardware to maximize throughput and minimize latency.
Role type
Senior Embedded Performance Engineer (ML Inference)
Builds
Runtime software for AI accelerators, ML kernels, and low-level control paths
Domain
Hardware/Software co-design for AI inference
Deliverable
production ML models
Required skills
C programming, Assembly (RISC-V), low-level architecture reasoning, memory behavior analysis, performance profiling, microbenchmarking, scheduling, synchronization, data movement
Preferred skills
GPU/ML accelerator experience, Rust, Python, simulator usage, modern ML inference workloads
Technologies
C, Assembly, RISC-V, ML kernels
Responsibilities
Write and optimize performance-critical ML kernels in C and Assembly; Build low-level control paths for scheduling and data movement; Write validation workloads and microbenchmarks; Profile and benchmark performance regressions; Collaborate with hardware, simulation, and compiler teams to drive architecture feedback
