AI Systems & Infrastructure Intern (LLM & GPU Workloads)

Micron Technology; Austin (TX); On-site; USD 34,000 - 44,000; Full time

Job summary

Micron Technology is seeking an AI Systems Software Engineering Intern to work with senior engineers on advanced systems software for LLMs and Agentic AI applications. You will profile and optimize AI inference/training workloads across GPU platforms and heterogeneous memory, interconnect, and storage systems.

The role emphasizes memory management techniques, benchmarking, and collaboration to develop AI workloads, publish results, and contribute to future platform designs.

Qualifications

• Currently pursuing a Master’s or Ph.D. in Computer Science, Computer Engineering, Electrical Engineering, or a related technical field.
• Demonstrated experience with AI systems, machine learning systems, computer systems research, or systems software development through coursework, research, or projects.
• Understanding of Large Language Models (LLMs), including transformer execution, attention mechanisms, KV cache behavior, batching, token-level latency, throughput, and memory performance considerations.
• Proficiency in Python and C/C++, with hands-on experience developing, debugging, and optimizing software in Linux environments.
• Experience using GPU-based performance analysis tools and at least one modern AI framework or serving stack, such as PyTorch, vLLM, TensorRT-LLM, NVIDIA Dynamo, or related technologies.

Responsibilities

• Develop and enhance systems software, profiling tools, and experimentation frameworks for LLM training, LLM inference, and Agentic AI workloads.
• Design, implement, and evaluate memory- and state-management techniques, including caching, tiering, compression, eviction, and lifecycle management for AI serving environments.
• Characterize and optimize AI workload execution across GPUs, CPUs, memory subsystems, storage, and distributed infrastructure, with a focus on latency, throughput, scalability, and resource utilization.
• Build benchmarking, simulation, and automation capabilities to evaluate data placement, migration, scheduling, and performance behavior across heterogeneous memory systems.
• Collaborate with engineering and research teams to develop representative AI workloads, analyze experimental results, and contribute to technical publications, intellectual property, and future platform designs.

Skills

Python

C/C++

AI systems

LLMs

Linux

GPU performance tools

Education

Master's or Ph.D. in Computer Science / Computer Engineering / Electrical Engineering

Tools

PyTorch

TensorRT-LLM

vLLM

NVIDIA Dynamo

Job description

Micron Technology is seeking an AI Systems Software Engineering Intern to work with senior engineers on advanced systems software for LLMs and Agentic AI applications. You will profile and optimize AI inference/training workloads across GPU platforms and heterogeneous memory, interconnect, and storage systems.

The role emphasizes memory management techniques, benchmarking, and collaboration to develop AI workloads, publish results, and contribute to future platform designs.