Research Intern - AI Compiler: Microsoft

This is a 12-week Research Internship providing a unique opportunity to contribute to the forefront of AI and high-performance computing. The project focuses on developing innovative techniques for automating and optimizing computational kernels, which are essential for accelerating modern AI systems.

Research Interns are paired with mentors and are expected to collaborate with other researchers, present findings, and contribute to the community.

Project Focus
The core project involves:

Pushing the capabilities of AI models for kernel rewriting and fusion.

Targeting optimization across various platforms, including GPU, Microsoft AI Accelerator (MAIA), and AMD platforms.

Rigorous benchmarking of new solutions against real-world workloads.

Making impactful contributions to open-source projects and academic literature.

Qualifications
Required Qualifications
Enrollment: Currently enrolled in a PhD program in Machine Learning, AI, Compiler, Algorithms, Systems, or a related area.

Experience: 1+ years of experience working with deep learning frameworks like PyTorch.

Application Requirements
Candidates must also submit a minimum of two reference letters for this position, as well as a cover letter and any relevant work or research samples.

Preferred Qualifications
Experience with AI compilers.

Experience with post-training models (Supervised Fine-Tuning, Reinforcement Learning, etc.).

Experience with writing GPU kernels in specialized languages (Triton, Compute Unified Device Architecture (CUDA), Gluon, Helion, etc.).

Demonstrated ability to develop original research agendas.