Member of Technical Staff, AI Engineering
Core
Architecting and optimizing large-scale custom model training, fine-tuning, and autonomous AI agents on multi-GPU clusters to power Micron's manufacturing stack.
Role type
Senior IC GPU Performance Engineer (AI/ML Systems)
Builds
Scalable ML systems, distributed training pipelines, and autonomous AI agents for semiconductor manufacturing
Domain
Semiconductor manufacturing, High-Performance Computing (HPC), Generative AI
Deliverable
production ML models
Required skills
GPU architecture (memory hierarchy, tensor cores, NVLink), distributed training (FSDP, DeepSpeed, Megatron-LM), CUDA/HIP programming, LLM fine-tuning (PEFT, LoRA), Python, C++, Kubernetes, CI/CD
Preferred skills
Ph.D. in CS/Statistics, Triton kernels, custom C++ extensions for PyTorch, HPC job schedulers (Slurm), multi-agent system design, computer vision/signal processing
Technologies
CUDA, HIP, PyTorch, TensorFlow, vLLM, TensorRT-LLM, LangChain, LangGraph, LlamaIndex, AutoGen, Kubernetes, Ray, Kubeflow, Jenkins, Docker, Git
Responsibilities
Architect large-scale custom model training and fine-tuning jobs; Optimize training throughput and memory efficiency; Design and develop autonomous AI Agents; Analyze and profile complex workloads to identify bottlenecks; Write and optimize high-performance kernels; Collaborate with Hardware Architects to define next-generation GPU features; Design and implement performance regression testing suites; Mentor engineers in parallel programming paradigms
Seniority
Senior, hands-on IC