I had a wonderful experience interning with the Visualization team at Apple this summer. I’m especially grateful to my manager, Yannick, and the entire team for being so supportive and great to work with. Over the summer, I focused on developing a fine-tuning approach informed by mechanistic interpretability insights, aimed at improving specific model capabilities while minimizing regressions on other skills learned during pre-training. I’m hopeful that we’ll be able to submit this work to ICLR 2026!

Although I was based in Boston, I was fortunate to have the chance to visit Apple Park for an HCI workshop on Responsible AI, hosted by the Human-Centered Machine Intelligence organization. It was an amazing opportunity to meet many Apple researchers and faculty members, and it made for an unforgettable first experience in the Bay Area!

Group photo at Apple Park!