Rob Roudani

• Founded and scaled a Responsible AI advisory practice enabling ethical, privacy-preserving, and explainable ML adoption across FinTech, HealthTech, and AdTech startups — reaching over 25M users across EMEA. • Architected and institutionalized AI governance frameworks aligned to EU AI Act and FTC standards, reducing client audit risk by 40% and accelerating compliant ML deployment by up to 30%. • Partnered with C-suite and boards to define Responsible AI charters, risk oversight structures, and fairness calibration programs — embedding AI governance into product and data organizations. • Delivered a Responsible AI toolkit suite (fairness dashboards, model cards, bias detection APIs) adopted by 10+ enterprise clients, standardizing lifecycle governance and improving explainability metrics by 25%. • Secured multi-million–dollar advisory partnerships and innovation grants to advance privacy-preserving ML (federated learning, differential privacy) and fairness-aware optimization frameworks. • Designed and launched executive learning programs training 500+ leaders and engineers in ethical AI operations and regulatory readiness, strengthening AI accountability culture at scale.

Algorithmic Fairness for Housing, Employment and Credit Facebook Ads - Core Ads strategy - Responsibility and Privacy - AI/ Machine Learning - AI Fairness & Ethics Machine Learning Responsible AI, Ads & Monetization Group | USA | 2021 – 2023 • Defined and led Meta’s Responsible AI strategy for Ads, in partnership with HUD,FTC and DOJ, negotiating fairness metrics, bias thresholds, and monitoring standards — embedding compliance into global ad delivery systems. • Directed 4 PMs and 30+ cross-functional engineers, data scientists, and policy experts to operationalize fairness, privacy, and transparency across ML pipelines powering billions of ad impressions per day. • Led development of the Variance Reduction System (VRS) — a reinforcement learning–based fairness optimization framework minimizing statistical variance in ad delivery, improving fairness metrics by 18% and performance trade-off • Institutionalized privacy-preserving ML (audience sampling, aggregation, differential privacy), cutting compliance risk 30% while maintaining predictive accuracy and low-latency delivery. • Championed the enterprise-wide AI Governance & Transparency roadmap, embedding fairness and explainability standards into the ad ecosystem — reducing measurable bias by 15% across protected segments. • Partnered with Legal, Policy, and Regulatory Affairs teams to establish ongoing fairness audit and accountability frameworks, now serving as the foundation of Meta’s Responsible AI operating model. • Launched the Housing Ads Classifier to ensure compliance with HUD/FTC guidelines, improving model precision by 12% and reducing reporting pipeline errors by 25% across sensitive categories. Scope & Impact: $XXB ad platform • Billions of daily impressions • managed 4 PMs, 30+ engineers • 55–70% fairness uplift • 30% compliance risk reduction • 99.9% reliability maintained.

Payments, Machine Learning for Credit, Risk & Fraud Credit Risk, Line Management & Authorizations | USA | 2019 – 2021 • Owned Amex’s global modernization credit line management and authorization platforms across 30+ markets — leading multi-year transformation to ML-driven, real-time decisioning systems managing millions of transactions daily. • Defined and executed the global AI product strategy for credit risk and line management, embedding advanced ML models to automate credit limit adjustments — cutting manual reviews 35% and improving utilization 17%. • Partnered with Data Science, Credit Risk, and Engineering leadership to deliver next-generation underwriting and portfolio optimization models — increasing approval precision 20% and enhancing customer lifetime value. • Re-architected data infrastructure and model pipelines integrating behavioral, bureau, and transactional features — lifting predictive accuracy by 15% and accelerating model deployment from quarterly to continuous delivery cycles. • Influenced executive-level portfolio strategy through data-driven insights and scenario modeling, shaping the company’s global approach to AI-enabled credit risk management and capital efficiency. Scope & Impact: Launched and scaled credit products in 30+ markets • Hundreds of millions of transactions daily • 30% precision lift • 65% manual review reduction • 25% model lift • Portfolio utilization +11%.

Founding team | Switzerland | 2015 – 2017 • Co-founded and led the product strategy for a SaaS digital-records platform, driving product vision, GTM strategy, and early-stage growth to secure product–market fit and SME adoption across Switzerland, Belgium, and France.

Intelligent Compute Group | France, Singapore, Germany, USA | 2006 - 2015 • Directed architecture and optimization of advanced microprocessor systems across multi-region R&D hubs, enhancing performance, efficiency, and reliability in high-throughput computing platforms.