This is for a 2025 summer internship with the following start dates: May - August or June - September for quarter system schools.

Work with a team of world-class researchers to design and implement compiler optimizations for AI accelerators targeting state-of-the-art LLMs (large language models). Develop compiler software for design space exploration to map AI applications to AI accelerators.Engage with other researchers to debate and discuss merits of existing and proposed compiler optimizations for AI accelerators.Prepare internal reports and presentations to summarize progress/blockers.Perform research and propose new concepts in compilers to exercise the computer architecture features.Implement the compiler optimizations to generate programs that exercise the computer architecture features.Evaluate the performance benefits using the software simulators across different use-cases and determine the hardware and software bottlenecks to improve performance.Performa cost-benefit analysis using the cost models and the simulator performance and propose changes to the hardware and software to improve the trade-offs.