May 2023 – Apr 2024 · 11 mos · New York, United States · Hybrid

• Deployed AI-based credit models with explainable AI (XAI) capabilities, reducing loan default rates by 20% and improving underwriting accuracy by 35% — saving $8M+ annually in credit loss exposure.
• Built first-party data activation pipelines using Salesforce and BigQuery, boosting product uptake 10% YoY across a 300,000+ member base.
• Introduced an NLP-powered voice-of-member analytics program, surfacing product friction points and delivering an 18-point NPS improvement within 12 months.
• Reduced data incidents by 65% within 6 months through a scalable governance model with data quality SLAs, role-based access controls, and metadata management.

Jun 2021 – Jan 2024 · 2 yrs 7 mos · New York, New York, United States

• Built a 15-person analytics team from scratch; deployed automated AI pipelines and audience intelligence tools delivering a 64% lift in ROAS across key client portfolios.
• Contributed directly to $30M in new business wins through AI-driven measurement storytelling in competitive pitches against holding-company agencies.
• Reduced audience build time by 70% through data partnerships with Snowflake, Resonate, and Zeta Global, improving targeting precision across campaigns.
• Delivered $21M in incremental revenue for a pharma DTC client by combining Crossix exposure data with patient panel research to optimize creative mid-flight — lifting brand awareness 40% and new patient starts 23%.
• Drove $14.7M in incremental oncology revenue within 90 days by integrating IQVIA prescription data with Crossix exposure logs, lifting TRx share 19%.

Mar 2020 – Jun 2021 · 1 yr 3 mos · New York, New York, United States

• Led AI and measurement strategy for L'Oréal, Macy's, and Audi — delivering $45M in optimized media spend across 12 global markets.
• Identified $12M in reallocatable budget through media mix remodeling for L'Oréal across 8 markets, improving blended ROAS 27% without incremental investment.
• Validated $8M in previously unattributed revenue for Macy's through geo-based incrementality testing, securing expanded measurement budget.
• Reduced media waste by 18% across the luxury portfolio through statistical overlap analysis and audience deduplication — delivering $6M+ in recovered efficiency.
• Achieved 30% improvement in cross-device attribution accuracy by integrating Google CM360, Nielsen, and LiveRamp identity resolution.
• Implemented first-of-kind HIPAA-compliant HCP data activation via clean room environments within Omnicom's measurement practice.

Jan 2018 – Mar 2020 · 2 yrs 2 mos · New York, New York · Hybrid

• Delivered 70% ROI uplift for Luxury Brands through bespoke MMM and MTA models across 9-figure media budgets.
• Developed a proprietary cross-channel attribution model for Chanel that quantified the halo effect of luxury print and OOH on digital conversion — influencing a $15M reallocation toward brand investment.
• Established the agency's first Bayesian forecasting capability, improving pacing accuracy 35% and reducing end-of-flight budget surplus/deficit by 80%.
• Reduced incrementality testing costs by 60% by building automated geo-lift and A/B testing infrastructure — creating a repeatable, agency-wide capability.
• Cut time-to-insight by 50% through data lake governance architecture and real-time campaign feedback loops.
• Reduced team attrition from 40% to under 10% within 18 months; contributed to 4 pitch wins representing $22M in new annualized revenue.

Jun 2000 – Dec 2017 · 17 yrs 6 mos · Phoenix, Arizona · On-site

• Built and led a 32-member global data science team across Phoenix, New York, London, and Singapore — one of the largest dedicated marketing analytics functions in financial services at the time.
• Pioneered AmEx's first CLV modeling framework, improving new cardholder quality scores 32% and reducing 12-month churn by 19%.
• Built real-time transaction anomaly detection models reducing fraud exposure by $22M annually at a false positive rate below 0.3% — a best-in-class benchmark.
• Deployed propensity models for cross-sell and upsell across the global card portfolio, driving $45M in incremental annual revenue.
• Built enterprise-grade credit risk models saving $15M+ annually in credit loss exposure and reducing false positives by 28%.
• Influenced $200M+ in annual media investment decisions through multi-country Marketing Mix Models
• Established AmEx's first A/B testing infrastructure, running 120+ annual experiments and embedding data-driven decision-making across marketing, product, and risk.