Team Lead, Core Actor ML (Trust & Safety)

I lead the Core Actor ML team, leveraging advanced AI and machine learning techniques to ensure account safety across Reels, Feed, Explore, Search, Stories, and Suggested Accounts. My work focuses on building robust and transparent classification systems that protect a community of 3 billion users.

Key Initiatives & Impact:

Recommendation Status Transparency & Appeals: Spearheaded the backend implementation of industry-first transparency tools within "Account Status." This system provides users with direct insight into recommendation eligibility. I am a co-inventor on the patent for this technology, which received global media coverage for setting a new standard in platform transparency.

High-Scale Enforcement: Architected and deployed Viral Account Enforcement to rapidly review accounts receiving outsized engagement. This system ensures near-100% policy alignment for Instagram’s most visible and highly-reported accounts.

Threads Launch & Spam Mitigation: Played a critical leadership role during the launch of Threads by rapidly shipping a scalable fix to mitigate a post-launch spam surge. This intervention resulted in an 85% reduction in spam reports, stabilizing the platform during its most critical growth phase.

Adversarial “Porn Bot" Mitigation: Neutralized scripted abuse of the Instagram Stories ‘like’ button by shipping a highly precise, adversarial-resistant solution. This resulted in a sustained 94% reduction in user reports on Stories and a 50% decrease in global push notification reports.

Teen Safety & Age-Appropriate Content: Led the implementation of industry-leading features to restrict teen exposure to age-inappropriate accounts. This involved building the core classification logic and integrating transparency and appeals workflows to ensure a safe, age-appropriate experience for younger users.