Senior Software Engineer, GPU Code Generation for Machine Learning Inference
Core
Develop advanced GPU inference software for TensorRT to accelerate state-of-the-art generative AI models for local and enterprise deployments.
Role type
Senior Software Engineer (GPU Compiler & Code Generation)
Builds
High-performance inference software for diverse generative AI models and GPU hardware configurations
Domain
AI Inference / GPU Computing / Compiler Development
Deliverable
production ML models
Required skills
C++, GPU architecture, parallel programming, performance analysis, compiler development, code generation
Preferred skills
CUDA, Triton, CuTe DSL, CUTLASS, MLIR, TensorRT, PyTorch, JAX, ONNX Runtime, vLLM, SGLang
Responsibilities
Design and implement efficient GPU code generation techniques balancing performance and latency; Develop compiler optimizations including operator fusion and dynamic scheduling; Profile workloads and analyze generated GPU code to identify bottlenecks; Develop robust software in C++ and Python for tooling and testing; Collaborate with GPU architects and compiler engineers to guide hardware and software design
Seniority
Senior, hands-on IC
