Naba Rizvi

It's pronounced "nub-buh rizz-vee"
She/her
born raised blooming

I am a multi-award winning CSE PhD Candidate at UC San Diego. I work on fine-tuning and evaluating LLMs, data collection and quality assurance for LLM benchmarking, and prompt engineering for LLM safety and alignment. I have first-author publications at FAccT, CHI, and CSCW, as well as ACL (in submission, with average 4/5 score), and served as an invited speaker at top conferences including NeurIPS, SigCSE, GHC, and Tapia. I even co created the curriculum for the human-centered AI course at UCSD. My research has been featured in both national and international media, reaching 21 nations, and even earned me a commendation letter from former VP Kamala Harris!

Selected Awards: 2024 J&J Lime Connect Finalist, 2023 CS Hero Award, 2020 NCWIT Collegiate Award, 2019 Adobe Research WiT Award, 2018 Google Women Techmakers Award, 5x Hackathon Winner (First Place: 2)

Collaborators: Dr. Nedjma Ousidhoum, Dr. Andrew Begel (Carnegie Mellon University)

Formerly at Google Health AI (2022), Microsoft Research (2020), Adobe Research (2019)

Research Interests: AI evaluation, benchmarking, safety, and alignment

Contact Me: [firstinitial][lastname]@gmail.edu