Intern - Memory Performance and Power Modeling
About The Position
Micron Technology is hiring an intern to analyze AI/ML workloads and evaluate next-generation computing and memory architectures. Working with system architects and performance engineers, the intern will develop methodologies to characterize memory behavior, generate workload traces, and identify performance bottlenecks across AI training and inference applications. Work will examine how large language models, recommendation systems, vision models, and multimodal AI applications interact with CPUs, GPUs, NPUs, memory subsystems, and interconnects. The role uses simulation, profiling, and data analysis tools to investigate performance bottlenecks, memory behavior, bandwidth requirements, and workload scalability. Its focus is memory trace-based workload analysis to inform architectural decisions for future memory technologies and AI platforms. The intern will work with real-world AI workloads, trace collection tools, and analytics frameworks to study memory access patterns, bandwidth demands, cache efficiency, data movement, and system bottlenecks.
Requirements
- Pursuing an MS or PhD in Computer Engineering, Computer Science, Electrical Engineering, Data Science, or a related field.
- Coursework or research experience in computer architecture, machine learning or deep learning, parallel computing, operating systems, data structures and algorithms, and performance analysis.
- Programming experience in Python and C/C++.
- Strong analytical, problem-solving, and quantitative reasoning skills.
- Excellent written and verbal communication skills.
Nice To Haves
- Experience with AI frameworks such as PyTorch, TensorFlow, JAX, or ONNX.
- Familiarity with LLMs, transformers, diffusion models, recommender systems, or vision models.
- Experience with workload profiling, tracing, or performance analysis tools.
- Experience with simulation, modeling, or architecture research.
- Experience with an AI code assistant.
- Experience with performance profiling tools such as Nsight, VTune, perf, rocprof, or similar tools.
- Knowledge of GPU, NPU, TPU, or accelerator architectures.
- Understanding of memory systems, including cache hierarchies and DRAM.
- Experience with architecture simulators such as Gem5, Accel-Sim, or similar tools.
- Familiarity with Linux development environments and data analysis frameworks.
Responsibilities
- Analyze AI/ML workloads, including large language models, recommendation models, computer vision, multimodal, and generative AI applications.
- Collect, process, and analyze memory traces from AI training and inference workloads.
- Characterize memory access patterns, locality, reuse behavior, bandwidth utilization, and latency sensitivity.
- Characterize compute, memory, storage, and interconnect behavior across AI training and inference workloads.
- Develop workload traces, performance models, and architecture simulation environments.
- Investigate system bottlenecks related to memory capacity, bandwidth, latency, cache utilization, and communication overheads.
- Evaluate architectural tradeoffs across CPUs, GPUs, NPUs, accelerators, and memory technologies.
- Create automation frameworks for workload execution, data collection, and results analysis.
- Validate simulation and analytical models using benchmark and hardware measurements.
- Present findings through technical reports, visualizations, and architecture reviews.
Benefits
- Choice of medical plans.
- Choice of dental plans.
- Choice of vision plans.
- Programs to protect income in case of illness or injury.
- Paid family leave.
- Paid time off.
- Paid holidays.
- Additional compensation may include benefits, bonuses, and equity.
Stand Out From the Crowd
Upload your resume and get instant feedback on how well it matches this job.
What This Job Offers
Job Type
Full-time
Career Level
Intern
