Software Engineer, GPU/AI/ML

What you'll do

• Own the AI software stack by establishing best practices from low-level GPU kernels to large-scale distributed systems.
• Improve training, post-training, and inference performance for LLMs and autonomous AI workloads.
• Co-design hardware and software by providing GPU architecture input and developing software for new accelerators.
• Optimize deep learning and open-source LLM inference/training kernels using HIP and CUDA.
• Diagnose performance bottlenecks in multi-GPU, distributed settings using profilers like ROCm Profiler and Nsight.
• Implement and optimize post-training techniques including SFT, RLHF, and GRPO.
• Mentor other engineers and influence technical direction across teams to scale the AI software strategy.

What we're looking for

• Bachelor's in Computer Science, Computer Engineering, Electrical Engineering, or equivalent experience.
• Master's degree (preferred); PhD (plus).
• Expert-level modern C++ and experience designing large, performance-critical systems.
• Strong grasp of GPU architecture, memory hierarchy, and kernel optimization using HIP or CUDA.
• Hands-on experience with ROCm, CUDA ecosystems, or ML framework cores like PyTorch, TensorFlow, or JAX.
• Experience diagnosing bottlenecks using profilers like ROCm Profiler and Nsight in multi-GPU, distributed settings.
• Deep understanding of transformers, attention, and the full model lifecycle including SFT, RLHF, and GRPO.
• Substantial professional experience in software development within performance-critical environments (preferred).
• Working knowledge of RTL design and Verilog/SystemVerilog for hardware-software co-design (preferred).