Founder at Olelo Intelligent systems pvt ltd

Bengaluru Area, India
Mumbai, Maharashtra
Bengaluru Area, India

San Francisco Bay Area

San Francisco Bay Area







Master of Mathematics, Computer Science
University of Waterloo

B.E., computer science
Birla Institute of Technology and Science, Pilani
Join Prashant and 10k+ other professionals on
Tal is the app where bosses meet engineers & builders directly, without middlemen
Download Tal BossFounder at Olelo Intelligent systems pvt ltd
I work on Artificial Intelligence helping machines infer and interpret context from as few inputs as possible. I firmly believe that strong AI is possible only with significant improvement in our current ability to model and infer context. Most of current research is focussed on Semantic Intelligence, building sophisticated knowledge graphs and inference algorithms to figure out precise and highly granular context from content(videos, images but mostly textual). I have extensively worked on almost every problem in NLP ranging from POS tagging, chunking to word sense disambiguation, entity recognition, sentiment analysis, document classification, review mining, text summarization. Most of this work has been inducted into Cogmeta, our audience intelligence platform to be launched soon. I also seriously dabble into alternative models of computation and hardware efficient for semantic and cognitive computing. I also work on large scale machine learning algorithms specifically real time regression, classification and clustering algorithms/techniques for massive data sets. I am primarily interested in techniques that employ high degree of parallelization and randomization to obtain approximate solution in real time. Specialties: Deep learning,Behavioral Targeting, Distributed Machine Learning Algorithms, Classifiers/Regression Algorithms for Web Scale Data, Advanced NLP Algorithms (Named Entity Recognition, Sentiment Analysis, Relationship Extraction), Hadoop, NoSQL Databases (Cassandra/HBase), Cloud Computing, Recommendation Engines, Distributed Computing, Statistical Data Visualization, Neuromorphic models of computation
India
claude
codex
deepseek
github copilot
ChatGPT
Microsoft Copilot
Group Engineering Manager at Microsoft, Sep 2020 - Present · 6 y, Bengaluru Area, India, Redefining customer engagement. Revolutionary Audience Intelligence Platform powered by years of research in semantic and cognitive computing helping marketers to move away from campaigns to "continuos engagement". We are currently in private beta mode.
Founder at Olelo Intelligent systems pvt ltd, Oct 2018 - May 2020 · 1 y 7 mo, Mumbai, Maharashtra, Building privacy first, secure, enterprise smart voice assistant for enterprises.
Founder at Interior Point Analytics, Dec 2011 - Feb 2015 · 3 yrs 3 mos, Bengaluru Area, India, Interior Point Analytics is researching and building next generation of personalization and recommendation technology which will revolutionize the way content is produced, published, curated, consumed, recommended and shared.
Principal Analytics Engineer at Exponential, Jun 2009 - Nov 2011 · 2 yrs 6 mos, San Francisco Bay Area, Exponential® is a technology-enabled media services company that offers innovative online advertising products and services through its subsidiaries Tribal Fusion® (ad network), EchoTopic® (contextual advertising), LeadGenuity®(lead generation), and FullTango®(performance marketing). As a member of the core analytics team, I work on several challenging analytical problems spanning various fascinating areas such as game theory, combinatorial auctions, statistical regression and prediction, mathematical optimization, and natural language processing.
Algorithms Engineer at Efficient Frontier, Oct 2006 - Jun 2009 · 2 yrs 9 mos, San Francisco Bay Area, Researched and implemented statistical models and optimization algorithms to manage multimillion dollar search engine marketing campaigns. Most of my work at Efficient Frontier was focused on building statistical models which computed reliable estimates from extremely sparse and noisy data. Designed hierarchical statistical models for estimating rates of rare events such as clickthrough rate(CTR), conversion rate at multiple resolutions in hierarchical sparse data Proposed and Implemented reliable heuristics to identify and deal with outliers in extremely sparse data Worked on NLP tools that mine semantic similarity between two keywords, two URLs, or keyword and URL
Master of Mathematics, Computer Science, University of Waterloo
B.E., computer science, Birla Institute of Technology and Science, Pilani