Sanjana's headshot

Hey, I'm Sanjana!

Thanks for checking out my website! I'm a CS and AI student @ MIT and am super excited about getting under the hood with AI models, particularly in the research area of interpretability.

I would like to build AI systems that are transparent and responsible. I'm a fervent believer that building superintelligent AI does not inevitabily push humans out-of-the-loop. But I also believe that humans will be out-of-the-loop unless we (1) build mechanisms that help us to monitor and intepret AI (2) craft thoughtful regulations and policy around AI safety and (3) think deeply about what we want a future with superintelligent AI to look like. I'm currently an undergraduate researcher in the Decentralized Information Group @ MIT's CSAIL under Dr. Lalana Kagal, where I'm applying mechanistic interpretability methods to build a confidence-aware question answering system. I was previously an undergrad researcher at Prof. Ila Fiete's lab intersecting evolution strategies with gradient-free LLM fine-tuning. I've also previously interned at Amazon (Ring), Nasdaq, and have also researched at MIT's Media Lab, Oakland University, and the University of Michigan.

Outside of the above, I'm a huge advocate for women in STEM! We belong here. I also love creative writing, long distance running, (tart) frozen yogurt, and all music by Taylor Swift.

Feel free to reach out to me to chat about any of the above and check out my projects!