2027 PhD AI Training Systems and Performance Engineer Intern/Co-Op
Core
Optimize large-scale AI training and fine-tuning workloads on AMD Instinct GPUs, developing tooling to improve scalability, efficiency, and developer productivity.
Role type
PhD AI Training Systems and Performance Engineer Intern
Builds
AI training workloads, performance analysis tools, and optimization workflows for AMD GPU platforms
Domain
High-performance computing, AI systems software, GPU architecture
Deliverable
production ML models
Required skills
Python, C++, PyTorch, JAX, TensorFlow, distributed training systems, GPU performance optimization, profiling tools (PyTorch Profiler, ROCm Profiler, VTune, Nsight), transformer architectures, LLM training techniques
Preferred skills
ROCm, HIP, Triton, GPU kernel optimization, publications in AI/HPC/Computer Architecture
Technologies
AMD Instinct GPUs, ROCm, PyTorch, JAX, TensorFlow, vLLM, SGLang, MPI, NCCL/RCCL, OpenMP
Responsibilities
Bring up new training workloads and foundation models, profile and analyze AI workloads to identify bottlenecks, develop tools for automated training setup and debugging, implement optimization strategies for throughput and memory efficiency, collaborate on evaluating emerging AI models and distributed training techniques
Seniority
PhD student, research-focused IC
