Inference Optimization Intern – Performance Modeling

About the role#

The Institute of Foundation Models is seeking an intern to work on high-performance computing and efficient AI inference. You will work alongside researchers and engineers to optimize the execution of large-scale foundation models on next-generation NVIDIA GPU architectures. This role focuses on low-level GPU performance analysis, kernel optimization, and hardware-aware inference acceleration.

What you'll do#

• Develop analytical performance models for GPU kernels and inference workloads.
• Build and validate a simulator to estimate theoretical hardware performance limits.
• Identify performance bottlenecks in compute, memory, communication, and scheduling.
• Analyze GPU execution using NVIDIA Nsight Systems and Nsight Compute.
• Investigate PTX and SASS code generation to understand low-level execution behavior.
• Collaborate with researchers and engineers to optimize inference kernels for transformer-based models.
• Evaluate utilization of Tensor Cores, memory bandwidth, caches, and instruction pipelines.
• Design profiling methodologies for Hopper and Blackwell architectures.
• Document findings and provide actionable recommendations for performance improvements.

What you'll need#

• Currently pursuing a degree in Computer Science, Computer Engineering, Electrical Engineering, Artificial Intelligence, High-Performance Computing, or a related quantitative discipline.
• Experience with CUDA programming and GPU kernel development.
• Understanding of NVIDIA GPU architecture and memory hierarchy.
• Familiarity with performance profiling tools such as Nsight Systems and Nsight Compute.
• Experience with deep learning frameworks such as PyTorch or TensorFlow.
• Proficiency in C++, CUDA, and Python.
• Knowledge of PTX, SASS, roofline analysis, and hardware utilization metrics.

Location & details#

• Location: Sunnyvale, California
• Term: Fall 2026
• Modality: On-site
• Status: Full-time, paid internship