Ahmed Fathy
Trust & Safety Architect — Policy, Data & AI Systems
Summary
8+ years leading high-risk Trust & Safety escalations and building enforcement systems at global scale across the world's largest video and gaming platforms. Expertise in moderating harmful content, refining policies through data analysis, and designing AI-powered enforcement pipelines. Combines deep escalation experience with cybersecurity investigation skills and the ability to bridge operational needs with technical teams.
Experience
Trust & Safety Architect — Policy, Data & AI Systems
YouTube — Accenture
- •Handle high-volume, high-risk cases within SLA — processing complex enforcement decisions at scale while maintaining accuracy under tight time constraints on the video platform.
- •Serve as team SME — guiding junior analysts on high-impact enforcement decisions, maintaining quality and consistency across a high-volume caseload within SLA.
- •Analyze moderation data to identify emerging patterns and policy gaps, then communicate actionable insights to the Policy team for guideline improvements.
- •Collaborate with cross-functional SMEs on regular calibration sessions and ideation workshops, aligning enforcement strategies across shifts and regions.
Senior Content Moderation Specialist
Sony PlayStation — Social Element
- •Detected and acted on account compromise patterns — identifying credential stuffing, unauthorized access, and account takeover attempts across the gaming network.
- •Enforced policy against cheating software and unauthorized scripts — identifying third-party tools, automation, and mods that violated platform terms of service.
- •Identified grooming and child safety risks in user interactions — detecting predatory behavior and potential CSAM, escalating through proper channels.
- •Flagged phishing and scam attempts targeting the gaming community — including fake log-in pages, fraudulent giveaways, and impersonation schemes.
Certifications & Credentials
Featured Projects
Interactive dashboard tracking child safety OSINT data across NCMEC and IWF sources with YoY trend charts and SQL schema narrative.
Policy enhancement project targeting evasive video content. Reduced prevalence by 35% through trend analysis, policy redrafting, and cross-functional deployment.
Automated QR-code phishing detection pipeline for gaming chat platforms combining computer vision, domain intelligence, and headless browser heuristics.
AI-powered content moderation tool using Hugging Face Inference API for multi-class toxicity classification with policy enforcement engine and severity scoring.
Languages
Arabic (Native) — English (Professional Proficiency)