Prahitha Movva is an AI researcher and engineer focused on how language models are evaluated and deployed in practice. She has published work at NeurIPS, COLM, and ACL workshops, with research centered on diagnosing gaps between benchmark performance and real capability. Her applied work spans high-stakes deployment contexts including a DSSG fellowship with UNICEF and ML projects at Spotify, Santander Bank, and Predii. She leads the ML Industry Program at Cohere Labs and runs the Community Champion sessions for the Applied AI Research track at LevelUp Labs where researchers and engineers debate what's actually working in production. She’s all about making sure research ideas actually make sense for engineering teams.
