Intern - Memory Performance and Power Modeling
Micron TechnologyJob Title
Intern - Memory Performance and Power Modeling
Role Summary
Work with system architects and performance engineers to analyze AI workloads and evaluate next-generation computing and memory architectures. Develop workload-trace methodologies, performance models, and simulation environments to identify memory and system-level bottlenecks for AI training and inference.
Experience Level
Internship - early-career role intended for students or recent graduates; typically filled by graduate students (intern level).
Responsibilities
Primary responsibilities focus on memory-focused workload analysis, modeling, and validation.
- Analyze AI/ML workloads including LLMs, recommendation systems, vision, multimodal, and generative models.
- Collect, process, and analyze memory traces from training and inference workloads.
- Characterize memory access patterns, locality, reuse, bandwidth utilization, and latency sensitivity.
- Develop workload traces, performance models, and architecture simulation setups.
- Investigate system bottlenecks related to memory capacity, bandwidth, latency, cache utilization, and communication overhead.
- Evaluate tradeoffs across CPUs, GPUs, NPUs, accelerators, and memory technologies.
- Create automation frameworks for workload execution, data collection, and analysis.
- Validate models with benchmarks and hardware measurements and present results in technical reports and reviews.
Requirements
Must-have technical skills and abilities, plus preferred qualifications.
- Must-have: Programming experience in Python and C/C++.
- Must-have: Strong analytical, problem-solving, and quantitative reasoning skills; effective written and verbal communication.
- Preferred / Nice-to-have: Experience with AI frameworks such as PyTorch, TensorFlow, JAX, or ONNX.
- Preferred / Nice-to-have: Familiarity with LLMs, transformers, diffusion models, recommender or vision models.
- Preferred / Nice-to-have: Experience with workload profiling, tracing, or performance analysis tools (e.g., Nsight, VTune, perf, rocprof).
- Preferred / Nice-to-have: Experience with architecture simulation or modeling tools (e.g., gem5, Accel-Sim) and architecture research.
- Preferred / Nice-to-have: Knowledge of GPU/NPU/TPU/accelerator architectures and memory systems (caches, DRAM).
- Preferred / Nice-to-have: Familiarity with Linux development environments and data analysis frameworks.
Education Requirements
Pursuing an MS or PhD in Computer Engineering, Computer Science, Electrical Engineering, Data Science, or a related field. Relevant coursework or research includes Computer Architecture, Machine Learning/Deep Learning, Parallel Computing, Operating Systems, Data Structures and Algorithms, and Performance Analysis.
About the Company
Company: Micron Technology
Headquarters: Boise, Idaho, USA
Micron Technology is a global leader in memory and storage solutions, dedicated to transforming how the world uses information. The company offers a diverse portfolio of high-performance DRAM, NAND, and NOR memory products under the Micron and Crucial brands. With a commitment to customer focus and technological innovation, Micron drives advancements in artificial intelligence, 5G, and other data-centric applications, empowering users to learn, communicate, and progress.
