Jan 2021 – Sep 2025 · 4 yrs 8 mos · Menlo Park, California, United States · Hybrid

• Expert in modeling user journey within/outside Meta Apps using in-app and out-of-app signals to skyrocket user/product growth and engagement via ML driven product integrations for high volume of online traffic (1B per day)
• Founding Engineer on Threads Ranking. Drive user growth DAU/MAU by 20%+. Led Feature Engineering, Graph Building, Threads Integration in IG/FB via ML pipelines of 50+ data sources and multi-layer modeling system.
• Improving integrity within Threads, Friending, Messenger via LLM & ML Ranking to reduce unwanted, clickbait, non-rec content by 50%+.
• Design A/B experiments to grow FB social graph via friend recommendation system trained on 100B+ data points for a billion connections daily.

May 2020 – Aug 2020 · 3 mos · Menlo Park, California, United States

• Implemented Feature Extraction Platform to emit Dataswarm pipeline tasks. Achieved quadratic to linear space/time complexity reduction
• Developed Recommendation Engine by leveraging the item-to-item Collaborative Filtering technique. Achieved 20% higher overlap with Ad features
• Feature engineering using Ad content and User signals. Achieved 2% Normalised Entropy reduction in production models and10% increase in event calibration

Aug 2020 – Dec 2020 · 4 mos · Atlanta, Georgia, United States

• Course: CS 6250-O01 Computer Networks
• Developing course content with the professor/TAs
• Hosting sessions to go over projects
• Implementing autograder in Python
• Grading exams and projects

Aug 2019 – Dec 2020 · 1 yr 4 mos · Atlanta, Georgia, United States

• Cyber Forensics Innovation Laboratory
• Reverse Engineering Malware
• Infiltrating Botnet
• Performing Memory Forensics

Sep 2017 – May 2019 · 1 yr 8 mos · Bengaluru, Karnataka, India

• Extended the Spring MVC backend to facilitate high volume Visa Token Service (VTS) provision and replenish flows
• Replaced the existing RDBMS based Search Platform with document based Elastic search, resulted in 20x decrease in search time
• Implemented restful service and certificate management for authentication of Tokenization flow through Embedded Secure Element
• Worked with Cyber Security team to address vulnerabilities in VTS Transactional Flows and Configuration Platform

Jun 2017 – Aug 2017 · 2 mos · Rennes Area, France

• Research Intern with TAMIS team https://team.inria.fr/tamis/
• Wire-Speed Packing Detection is a necessity in the modern world as most of the malware writers use packing to obfuscate the malicious code in the executables
• Worked on improving the current state of the art methods for very fast test time analysis
• Built a packed executable filtering system to be used in routers for packet filtering
• Constructed a pool of features using metadata and entropy histograms
• Performed feature selection considering the contribution and computation cost of each feature
• Extracted features from binaries and used ML algorithms to build a detection model
• Found a reduction of 1-2% effectiveness can increase efficiency by 17-44 times
• Paper accepted in Elsevier's Journal of "Computer & Security" for August 2019

May 2016 – Jul 2016 · 2 mos · Dallas/Fort Worth Area

• Function Boundary Identification in stripped COTS binary is a critical step in many binary code analysis applications such as reverse engineering and malware analysis
• Used exception handling information that is present in the COTS binary for function boundary identification
• The approach did not require any machine learning or intensive computation unlike the current state of the art methods. The approach was completely automated and worked across all different levels of optimization and compilers
• Achieved state of the art results with 100% precision and around 99.87% recall.

May 2015 – Jul 2015 · 2 mos · Gurgaon, India

• Developed new metrics on non verbal cue analysis for content rating based on Monet’s existing platform
• Integrated the developed metrics and associated presentation charts into existing Monet code
• Added new gamification and user engagement statistics
• Developed a basic Video Recommendation System to improve Monet's user experience