Rhea Kapur
Research Assistant Stanford NLP Group at Stanford University School of Engineering
United States
Summary
👤
Senior🎓
Top SchoolRhea Kapur is a research-focused machine learning engineer with eight years of experience at the intersection of NLP, speech technology, and computational linguistics. Currently a Research Assistant in the Stanford NLP Group, she studies information density and evaluation metrics for image description while building diarization and ASR models for child speech and classroom settings. Her industry work includes reducing hallucinations by 95% in production ASR pipelines and introducing emotionally nuanced voice agents, and she has experience scaling hybrid distributed LLM training. Rhea’s academic work spans a noted COLING2020 paper on language evolution simulations and a dual background in Computer Science (AI) and Linguistics from Stanford, reflecting a rare blend of empirical rigor and product-minded engineering. Notably, she applies linguistic theory to practical speech systems, bringing accessibility-focused evaluation into mainstream image and audio generation research.
8 years of coding experience
4 years of employment as a software developer
Master of Science - MS, Computer Science (AI), Master of Science - MS, Computer Science (AI) at Stanford University School of EngineeringBachelor's degree, Computer Science (AI) & Linguistics, Bachelor's degree, Computer Science (AI) & Linguistics at Stanford University
Spanish, Russian, Hindi
Github Skills (15)
speaker9
diarization8
building-blocks8
speech-processing8
voice-activity-detection8
embedding7
speaker-verification7
speech7
overlapped-speech-detection7
change-detection7
pytorch6
speaker-diarization6
speaker-recognition5
deep-learning5
speech-recognition5
Programming languages (3)
Github contributions (5)
iRT-CS/ShallowMind
Dec 2018 - Jan 2021
Contributions:45 pushes in 2 years 1 month
rkapur102/allerscan
Feb 2021 - May 2021
Contributions:23 commits, 26 pushes, 1 branch in 2 months
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.