Staff Software Engineer, AI/ML Heterogeneous Systems
Core
Build the high-performance software stack for deploying LLMs and Generative AI models on RISC-V architectures, spanning compiler infrastructure to distributed runtime systems.
Role type
Staff Software Engineer, AI/ML Heterogeneous Systems
Builds
Production-grade inference stacks (vLLM, SGLang) and core compute programming models for RISC-V hardware.
Domain
Semiconductor / AI Infrastructure / Heterogeneous Computing
Deliverable
production ML models
Required skills
C++, Python, Heterogeneous Computing execution models, LLM serving internals, Compiler infrastructure (Triton/MLIR), Memory management, Distributed systems
Preferred skills
RISC-V Vector (RVV) extensions, CUDA C++, Open Source contributions (vLLM, SGLang, PyTorch, LLVM/MLIR) (via careerplan.io/jobs/R-101299-staff-software-engineer-aiml-heterogeneous-systems-at-sifive)
Technologies
vLLM, SGLang, Triton, MLIR, RISC-V, CUDA, ROCm, OpenCL
Responsibilities
Design and implement core compute programming models and device runtime daemons; Optimize memory-bound operations and KV-cache management for inference stacks; Develop high-performance kernels for RISC-V Vector and matrix compute units; Collaborate with hardware architects on future AI accelerator designs.
Seniority
Staff, hands-on IC with architectural influence
