Software Development Engineer
Core
Optimizing and developing deep learning frameworks for AMD GPUs to enhance kernel performance and enable training/inference at scale.
Role type
Senior IC software engineer (GPU deep learning frameworks)
Builds
High-performance GPU kernels, optimized deep learning frameworks (TensorFlow, PyTorch, SGLang), and distributed training/inference systems
Domain
AI/ML, GPU computing, compiler technology
Deliverable
production ML models
Required skills
C++, Python, Linux, GPU kernel development, deep learning frameworks, distributed systems, compiler integration, profiling and optimization
Preferred skills
SGLang, LLM optimization, compiler design (LLVM, MLIR), ROCm, heterogeneous cluster workloads, CUDA/HIP, GCN/CDNA architecture
