Mountain View, CA
July 2018 - present
July 2018 - present
- ML and AI algorithms for LinkedIn Communications and Notifications.
Mountain View, CA
Oct 2014 - June 2018
Blog Posts:
https://www.bloomreach.com/en/blog/2018/07/using-machine-learning-to-learn-page-templates.html
http://engineering.bloomreach.com/identifying-new-product-groupings-in-e-commerce/
http://engineering.bloomreach.com/grace-hopper-2015/
Oct 2014 - June 2018
- Developing ML algorithms to enhance the discoverability and relevance of content on e-commerce sites.
- Part of the Organic Search team. Working on algorithms to determine Related Searches and Related Products for a given product/category.
- Sample Projects completed at BloomReach:
- Canonicalization Pipeline: Determine unique URLs on e-commerce websites based on URL content
- Global Link Graph Optimization: Algorithms to determine a better linked structure on e-commerce websites based on certain requirements.
- Template Identification on e-commerce sites: Determine the unique set of templates on e-commerce sites from millions of web pages using machine learning.
- A/B Test Framework: Designed and lead the development of a better A/B test framework to study impact of BloomReach widgets on traffic to e-commerce sites.
Blog Posts:
https://www.bloomreach.com/en/blog/2018/07/using-machine-learning-to-learn-page-templates.html
http://engineering.bloomreach.com/identifying-new-product-groupings-in-e-commerce/
http://engineering.bloomreach.com/grace-hopper-2015/
Natick, MA
May 2011 - Sept 2014
May 2011 - Sept 2014
- Worked on Simulink Design Verifier, a toolbox which uses formal methods to statically detect errors like overflow, division by zero, and automatically generates tests to satisfy a range of coverage objectives.
- Added new capability to the toolbox to automatically generate test cases to satisfy relational boundary objectives, a major enhancement released in R2014b.
- Worked with intricacies of data types and type casting, compiler optimizations, code generation and IR frameworks, SAT solvers.
- Designed and developed a memory leak management tool for iteratively identifying new memory leaks. Voted the 2nd most popular tool among 110 submissions for tools to improve the quality of MathWorks products.
Research Assistant, Robot Vision Lab
West Lafayette, IN
Aug 2008 - May 2011
West Lafayette, IN
Aug 2008 - May 2011
- Proposed models for simulating large scale computer networks infected by malicious software.
- Designed and developed a graph theoretic framework to detect malicious activities in computer networks
Research and Technology Center
Intern
Pittsburgh, PA
Summer 2010
Intern
Pittsburgh, PA
Summer 2010
- Worked on a smart surveillance system being researched and prototyped at Bosch RTC.
- Developed an algorithm to automatically detect contextual anomalies by unsupervised modeling of spatio-temporal data from multiple sensors.
Project Research Associate
Hyderabad, India
Spring 2010
Hyderabad, India
Spring 2010
- Developed supervised and unsupervised learning based algorithms for identifying malicious activity in computer network data streams.
B.Tech (ICT)
2004 - 2008
2004 - 2008
- Proposed a sampling-resampling based Bayesian learning algorithm for learning mixture models for robust background subtraction in videos and image segmentation.