Ramesh Sarukkai Faculty Berkeley

Ramesh Sarukkai

Lecturer

ramesh.sarukkai@ischool.berkeley.edu
linkedin.com/in/sarukkai/

Focus

Technologist with deep interest in delivering intelligent solutions at scale. He has a track record of delivering impactful products & platforms at scale across multiple domains - social media, video/media, commerce/payments, speech and self-driving. Dr. Ramesh Sarukkai is currently the Chief Product and Technology Officer at Rate ( https://www.rate.com - A mortgage/FinTech company - #2 mortgage retailer in the US), and Founder of Prajna AI ( https://www.prajna.ai - GenAI solutions provider to financial sector).

Research areas

Data Science
Entrepreneurship
Information Retrieval & Search
Technology for Developing Regions

Biography

Dr. Ramesh Sarukkai is a seasoned technologist with over 20 years experience in the silicon valley. He has led groups to deliver on impactful projects including YouTube monetization at Google, Payments/commerce at Facebook, autonomous mapping at Lyft, as well as commerce/payments solutions at Braintree/PayPal. He is currently the Chief Product and Technology Officer at Rate Inc ( https://www.rate.com/ - top 5 mortgage retailer in the US), where he is leading FinTech transformation and innovating with intelligent automation  with the goal of making people's financial dreams come true. In addition, he is also the Founder and President of Prajna Inc ( https://www.prajna.ai/ ) -  a generative AI startup focused on bringing no-code solutions to various areas within the financial enterprise. Prajna is a part of the NVIDIA Inception group as well as a member of Google Startups Program. Current research focus is innovative enterprise applications of Generative AI technology.

His background is in computer science, with a industry focus on delivering products/platforms at scale to hundreds of millions of customers and driving billions of dollars of revenue to the business. With this experience, he has had the opportunity to bring advanced computing and machine learning technologies to significantly shape and improve real-world products across the different domains. 

He holds a PhD in computer science from the University of Rochester, a prolific patenter, and has authored a book on "Foundations of Web Technology" (Kluwer/Springer). He also participates in leading conferences as a author, panelist or invited speaker (e.g. 2017 NIPS workshop on Machine Learning for Intelligent Transport Systems, CVPR'18 Workshop on Autonomous driving CVPR'18 and more)

Currently Dr. Sarukkai's focus of research is delivering large scale innovation to a broad spectrum of domains (fintech, commerce, social, media). In particular, he is actively focusing on GPT models for enterprise and financial applications.

Sample research work:

  1. Invited talk on "Autonomous Mapping" at CVPR'18 Workshop on Autonomous Driving
  2. "Micro-perception Approach to Intelligent Transport", 2017 NIPS Workshop on Machine Learning for Intelligent Transportation Systems
  3. "Accurate Deep Direct Geo-Localization From Ground Imagery and Phone-Grade GPS", , Shaohui Sun, Ramesh Sarukkai, Jack Kwok, Vinay D. Shet, CVPR Workshops, 2018. 
  4. "YouTube monetization: creating user-centric experiences using large scale data" in Proceedings of the 23rd International Conference on World Wide Web.
  5. "Creating the Perfect Video Stream", Proceedings of second ACM SIGMM workshop on Social media, 2010.
  6. "Joint categorization of queries and clips for web-based video search", R Zhang, R Sarukkai, JH Chow, W Dai, Z Zhang, Proceedings of the 8th ACM international workshop on Multimedia information Retrieval, 2006.
  7. "Multimedia information retrieval: what is it, and why isn't anyone using it?", Keynote,Proceedings of the 7th ACM SIGMM international workshop on Multimedia Information Retrieval, 2005.
  8. "Video search: opportunities & challenges", Keynote, Proceedings of the 7th ACM SIGMM international workshop on Multimedia information retrieval, 2005.
  9. "How much is a keyword worth?", Special interest tracks and posters of the 14th international conference on World Wide Web, 2005.
  10. "Exploiting the Dynamic networking Effects on the Web", Special interest tracks and posters of the 14th international conference on World Wide Web 2005.
  11. "Link prediction and path analysis using markov chains", Computer Networks 33 (1-6), 377-386, 2000 (and co-presented at WWW conference).
  12. "Phonetic set indexing for fast lexical access", Ramesh Sarukkai & Dana Ballard, IEEE Transactions on Pattern Analysis and Machine Intelligence 20-1, 1998.
  13. "Word set probability boosting for improved spontaneous dialog recognition", Ramesh Sarukkai & Dana Ballard, IEEE transactions on speech and audio processing 5 (5), 1997.
  14. "Supervised networks that self-organize class outputs", Neural Computation (MIT Press), 1996.
  15. "Supervised self-coding in multilayered feedforward networks", IEEE Transactions on Neural Networks 7 (5), 1996.

Recent courses taught