hey, i'm nikhil!

i’m an undergraduate at uc berkeley studying electrical engineering and computer science, with a focus on machine learning and deep learning. i’m broadly interested in building intelligent systems, especially at the intersection of machine learning, robotics, systems, and quantitative finance.

before college, i spent a lot of time doing competitive math and qualified for the usa math olympiad three times and the aime seven times. i love applying problem-solving knowledge across various domains, whether it's tackling complex algorithms, designing systems, or exploring new research areas.

recently, i worked as a quantitative trading intern at sig, where i explored high-frequency order-flow data and trained machine-learning models to study order execution and price impact. before that, i worked on generative ai at scale ai and built a full-stack academic platform at hermis.

at berkeley, i’m a researcher in the robotic ai learning lab within bair, advised by sergey levine and william chen. my work explores language-conditioned robot policies and ways to incorporate corrective human feedback into policy learning.

i also enjoy digging into how machine-learning systems work internally. some of my recent projects include attribution patching for multilingual t5 models and predictive modeling with messy real-world data like finance. i'm trying to do some cooler and more fundamental projects in the systems space as well.

i'm also pretty interested in entrepreneurship and i'm always building something on the side. if you have any cool ideas, let's chat!

in my free time, i like going outdoors and playing sports like basketball and soccer (i also love tennis and badminton). i'm on one of berkeley's bollywood fusion dance teams (ucb azaad!), and i enjoy following fantasy sports. when i'm not being active, you can find me playing strategy and board games, reading sci-fi novels, or trying new boba shops. and of course, i love spending time with my adorable german shepherd, cookie 🐕 !

view my work →