Experience
Industry
At Amazon, I worked on efficient algorithms and distributed methods for pre-training and fine-tuning LLMs. Before that, I worked in the personalization org at Yahoo! on developing machine learning models for Entity Matching, Entity Detection and built a Knowledge Graph from scratch to power Yahoo! websites. Prior to that, I worked at ThoughtWorks as a Software Developer.
| Nov 2024 - Present | Research Scientist, Meta, Menlo Park |
| Apr 2021 - Nov 2024 | Applied Scientist, Amazon AWS AI, Santa Clara |
| Apr 2020 - Apr 2021 | Applied Scientist, Amazon Alexa AI, Sunnyvale |
| Jul 2011 - Jul 2013 | Software Engineer, Yahoo!, Sunnyvale |
| Jun 2008 - Jul 2009 | Application Developer, ThoughtWorks, Bangalore |
Internships
During my PhD internships, I have worked on search relevance, learning to rank, recommender systems, text and document analysis and user behavior clustering.
| Summer 2017 | Applied Scientist Intern, Amazon AI, Palo Alto |
| Summer 2016 | Research Intern, Adobe Research, San Jose |
| Summer 2015 | Research Intern, Microsoft Research (Cloud Information and Services Lab), Mountain View |
| Summer 2014 | Research Intern, LinkedIn (Search Relevance team), Mountain View |
Academic Experience
| (2014 - 2019) | PhD, Computer Science, UC Santa Cruz |
| (2009 - 2011) | Masters, Computer Science, Georgia Tech |
My PhD research focussed on designing efficient distributed stochastic optimization algorithms for large-scale machine learning problems such as Extreme classification (multi-class/multi-label with large # labels and data points), Extreme clustering (mixture models with large # classes and data points) and Scalable robust learning to rank (large # of users and items). I have experience in implementing parallel, asynchronous optimization algorithms in distributed memory settings (using C++, MPI) for a variety of machine learning models.
During my Masters, I worked in the Sonification Lab of Georgia Tech, with Prof Bruce Walker in developing non-traditional interfaces for Human Computer Interaction. Prototyped tools and contributed to research of Auditory Menus. Used affect-detection libraries (machine learning to detect mood of the drivers) as part of a Next-Gen In Vehicle Interface project. Participated in lab-demos, poster-presentations and co-authored four publications. During my Undergrad, I worked on developing efficient Caching Algorithms for Location-Dependent Data.
Services
- PC member of AAAI Conference on Artificial Intelligence (AAAI)
- Reviewer of Neural Information Processing Systems (NeurIPS)
- Reviewer of International Conference on Machine Learning (ICML)
- Reviewer of International Conference on Learning Representations (ICLR)
- Reviewer of International Conference on Artificial Intelligence and Statistics (AISTATS)
- Reviewer of SIAM International Conference on Data Mining (SDM)
- Reviewer of Conference on Learning Theory (COLT)
- Reviewer of Conference on Uncertainty in Artificial Intelligence (UAI)
- Reviewer of Journal of Machine Learning Research (JMLR)
- Reviewer of IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI)
- Senior Member, IEEE