Jan 2025 – Present · 1 yr 5 mos

• Lead, Messaging (xMS/RCS) & Telephony Abuse: Directing end-to-end fraud and abuse mitigation strategies for global messaging products.
• Cross-Functional Security Architecture: Partner with XFN teams to design multi-layered anti-abuse protections, leveraging reputation signals (Phone Number, User, IP) and providing strategic counsel on counter-abuse frameworks.
• ML & Heuristic Productionization: Spearhead the selection, training, and deployment of machine learning models and heuristic rule engines to detect and neutralize evolving spam patterns at scale.
• Infrastructure & Financial Safeguarding: Lead independent, high-stakes projects to defend Google’s infrastructure from exploitation, effectively preventing large-scale financial fraud and protecting millions of users from digital harm.

Oct 2023 – Jan 2025 · 1 yr 3 mos

• Working with the User identity and Google telephony team on 'International Revenue Share Fraud'
• Leading an antispam abuse team on SMS fraud
• Working closely with multiple stakeholders (including directors and senior managers) in developing protections
• Saved $250m of SMS cost in FY24
• Writing automated rules (Python) and unit test for protections
• Analyzing large complex databases to develop new signals and metrics
• Awards: Received 25+ peer and spot bonuses for extraordinary work

Nov 2021 – Oct 2023 · 1 yr 11 mos

• Project 1: ML Model to flag high risk chrome extensions
• Identified 24k+ policy violating extensions by creating a machine learning model with an AUC score of 98%, that predicts the likelihood of spam and abuse in published extensions
• Project 2: Signal Mining: Unraveling Trends in Unstructured User Reports to Combat Abusive Extensions
• Actioned on 1k+ extensions, protecting 11m users from spam and abuse
• Streamlining Enforcement: Automated Pipeline Development for 80% Reduction in Manual Analysis and 74% Faster Enforcement
• Project 3:Securing Innovation: Data-Driven Identification and Resolution of Security Loopholes in Product Launches
• Identified and resolved multiple security loopholes, preventing exploitation by bad actors. Collaborating with engineering and product teams, we fortified defenses and optimized code review processes for enhanced platform security
• Conducted multiple A/B testing and experimentation to measure the performance of various defence models across the Chrome platform
• Protected 20M active users from potential spam and abuse
• Project 4: New Feature launches with the Product and Engineering team
• Launched 'Featured Badge' workflow to help end users in identifying high-quality items and putting them across search index query and recommendation algorithm, there by taking the high quality items install rate to 70%
• Project 5: Driving efficiency through data analytics and GenAI (LLM Models)
• Reduced manual review volume by 29% and realized $1.7m of annual savings by implementing system and process level changes
• Reduced developer appeals rate by 30%
• Created a code review prompt extension with the help of Gemini(Google LLM) to help manual reviewers in reviewing complex codes