Research Aide - LCF - Hossain, Md Muzakker - 3.13.26.
About The Position
Modern deep learning workloads on GPUs and emerging accelerators generate highly complex performance traces, making manual analysis of profiler tool outputs time-consuming, error-prone, and inaccessible to non-experts. These trace files contain millions of heterogeneous events spanning Python execution, kernel launches, memory operations, and distributed collective communication, but lacks a structured, interpretable abstraction for automated performance diagnosis. This project explores the design of an AI-assisted, agent-based framework for analyzing Torch/iProf trace files. The framework decomposes the analysis into specialized agents responsible for trace parsing and normalization, kernel-level performance characterization, host–device synchronization and stall detection, and distributed communication analysis. By transforming raw profiler traces into a structured intermediate representation and applying agentic reasoning on top of it, the system aims to automatically identify performance bottlenecks. The intern will work at the intersection of machine learning systems, performance engineering, and AI-driven tooling, contributing to scalable trace parsing, heuristic inference of CPU–GPU relationships, and natural-language generation of actionable performance insights. The outcome of this work will be prototype a “AI profiler analyzer” capable of producing human-readable performance summaries and optimization recommendations from raw profiler traces, with relevance to large-scale training on HPC and accelerator-rich systems. Education and Experience Requirements The entirety of the appointment must be conducted within the United States. Applicants must be: o Currently enrolled in undergraduate or graduate studies at an accredited institution. o Graduated from an accredited institution within the past 3 months; or o Actively enrolled in a graduate program at an accredited institution. Must be 18 years or older at the time the appointment begins. Must possess a cumulative GPA of 3.0 on a 4.0 scale. If accepting an offer, candidates may be required to complete pre-employment drug testing based on appointment length. All students remain subject to applicable drug testing policies. Must complete a satisfactory background check.
Requirements
- The entirety of the appointment must be conducted within the United States.
- Currently enrolled in undergraduate or graduate studies at an accredited institution.
- Graduated from an accredited institution within the past 3 months; or
- Actively enrolled in a graduate program at an accredited institution.
- Must be 18 years or older at the time the appointment begins.
- Must possess a cumulative GPA of 3.0 on a 4.0 scale.
- If accepting an offer, candidates may be required to complete pre-employment drug testing based on appointment length. All students remain subject to applicable drug testing policies.
- Must complete a satisfactory background check.
Responsibilities
- contributing to scalable trace parsing
- heuristic inference of CPU–GPU relationships
- natural-language generation of actionable performance insights
Benefits
- comprehensive benefits are part of the total rewards package
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
Education Level
No Education Listed
Number of Employees
1,001-5,000 employees
