Resume

First employee at a startup that evolved from a consumer app into a consumer insights platform using longitudinal diary studies. Helped shape the business direction and established the research methods, analytical standards, and measurement capabilities to define and test its thesis, personally delivering the supporting analysis and systems. Multiple customer pilots are underway ahead of go-to-market.

• Enabled the pivot from a consumer app to consumer insights. Demonstrated the opportunity through analytical work and built the technical and methodological foundation for the transition, collaborating with the CEO who led the business decision.
• Established research standards for AI-assisted consumer insights. Developed standards spanning adaptive data collection, qualitative analysis, evaluation, and customer reporting. Specified how to distinguish participant evidence from study and model artifacts, audit AI-generated findings, and bound claims to what the data support; extended this foundation into the digital-twin offering.
• Established measurement of the business thesis and unit economics. Defined customer lifetime value metrics before company founding, translating lifecycle economics into testable hypotheses. Built a warehouse covering costs, user behavior, and acquisition performance, with an MCP server giving business users natural-language access to metrics.
• Developing a digital-twin offering to package qualitative research at scale. Extended earlier research methods into a participant-modeling prototype, contributing the method, part of the design, and nearly all implementation in collaboration with a colleague who proposed the concept and supplied part of the roadmap. The prototype combines behavioral models with AI-assisted answers grounded in participant evidence.
• Developed and evaluated behavioral clustering methods. Combined semantic embeddings, density-based and hierarchical clustering, and sequence analysis; used randomized baselines and synthetic corpora with known patterns to assess pattern recovery and false discoveries, identifying distortions from correlated product attributes and sparse text-derived features.
• Built a prototype AI-assisted qualitative-analysis engine. Combined clustering and LLM-assisted coding with reproducible processing, source traceability, and human-review gates for codebooks and research claims.
• Established the production foundation and accelerated delivery of new services. Independently developed typed Python services, Dagster orchestration, a dbt/DuckDB warehouse, Quarto reporting, cloud deployment, and CI/CD. Built a reusable services framework with 13+ modules that reduced new-service development time to under a day, demonstrated by MCP and SKAdNetwork ingestion services.
• Implemented privacy controls for permissioned consumer data. Built consent handling, retention rules, and access boundaries into services to address GDPR and CCPA requirements.
• Designed a Bayesian product-experimentation system. A two-part hurdle model separates participation from intensity, with configurable priors, minimum-detectable-effect and credibility gates, and explicit checks against overlapping interventions.
• Prototyped AI conversation guidance via soft state machines. Replaced rigid scripted flows with probabilistic state transitions to adapt conversations to user responses.
• Ran Meta user acquisition end-to-end. Expanded into new Comscore geographies, tested acquisition creative systematically, and fixed upper-funnel conversion leaks.

Founding engineer at a fintech startup, delivering a production-ready rules engine and API layer for real-time-payment fraud detection.

• Built automated, policy-driven fraud workstreams for real-time payments. A service that generates Drools DRL rules dynamically from JSON definitions lets financial institutions encode their own policies without engineering involvement; a companion classification service evaluates complex multi-rule transactions in under 10 milliseconds.
• Delivered the backend API gateway and integration layer. Built in Python with PostgreSQL, connecting the front end to the rules engine and leaving the system ready for client onboarding.

Sole data scientist at a 20-person user-created games platform; converted from contractor to full-time based on results. Built the company’s measurement capability and advised the CEO on product decisions.

• Established product measurement across the creator and player marketplace. Defined telemetry, selected and implemented Amplitude and Statsig, and built pipelines to measure game audience development and player retention.
• Evaluated engagement with AI-assisted game creation. Combined product instrumentation, experimentation, and behavioral analysis to assess sustained engagement.
• Informed a strategic pivot through engagement analysis and experimentation. Used an engagement framework, A/B tests, and bandit-style tests to identify limitations in the mobile product and recommend a new direction. Following the engagement, the company executed the pivot and announced $56M in new funding.
• Improved onboarding completion and engagement. Used A/B and bandit-style tests on onboarding and game design to simplify the first-user experience.
• Connected core-user feedback to the product roadmap. Identified high-value users through behavioral analysis and used a user council to collect their feedback.

Promoted four levels in three years (Manager → Senior Director), ultimately leading a 20+ person organization spanning business intelligence, product and marketing analytics, and data science — 10 direct reports including two managers and approximately 12 indirects — while on the leadership team of a 250-person organization. Led analytics across a portfolio of in-app-purchase mobile games, connecting acquisition, player behavior, purchasing, retention, and CLTV to product, marketing, and investment decisions.

• Aligned cross-functional stakeholders around a revised launch forecast. Convened a working group and built a transparent, bottom-up cohort model with assumptions owned by contributing teams. The group’s shared assessment informed leadership’s cancellation of the original ~$100M budget and revision of launch strategy.
• Built a marketing analytics team that turned re-engagement programs from negative returns to 200% ROI. Built trust with marketers by understanding their constraints and developing analysts’ business fluency. Secured agreement to form the team, hired its manager, and delegated ownership of measurement and continuous improvement, using a shared marginal-return standard to guide campaign investment.
• Grew revenue by 5%+ through portfolio-level budget optimization. Guided acquisition budget allocation using differences in marginal incremental returns across channels. Supporting BI systems connected campaign spend, incremental revenue, and cohort lifetime value to estimate returns on additional spend for a budget in the tens of millions.
• Led causal marketing measurement and media mix modeling. Quantified channel contribution and identified more than $1M in spend-reallocation opportunities to improve return on advertising spend.

• Cut marketing analysis effort in half. Led a cross-functional modernization of the BI practice, migrating from Hive/Hadoop to a scalable cloud data warehouse with analytics-friendly schemas.
• Maintained iOS acquisition budgets through Apple’s App Tracking Transparency changes. Built SKAdNetwork pipelines and adapted measurement methods to the new attribution constraints.
• Designed the data policy for consumer mobile titles under GDPR and CCPA, setting how player data was collected, retained, and used across the portfolio.

• Expanded product analytics through joint prioritization with product managers. Established a shared backlog where analysts and PMs proposed investigations and prioritized work by expected product impact, creating room for analyst-led hypotheses alongside product requests.

Led data science teams working across product and engineering. Partnered with the CTO to define priorities, scope, and acceptance criteria for search and content systems.

• Launched a stalled related-search system within two months. Redefined acceptance criteria, automated the build, and removed stale data from an NLP system after six months of prior development, delivering a product worth more than $500K in annual revenue.
• Reduced vertical search site build time from three months to three days. Directed development of a tool automating index construction for narrow-topic search sites.
• Introduced Bayesian A/B testing where no testing capability existed. Built the company’s first mechanism for measuring product improvements.

• Established software delivery practices for the data science team. Adopted Scrum and built GitLab CI/CD pipelines to increase delivery cadence.
• Led a keyword bidding automation initiative. Delivered automated auction data extraction and bid adjustment within two months.
• Led deployment of a machine-learning control for search safety. Used a word-vector classifier to identify risky queries and block prohibited or off-brand content, including content related to self-harm.

Selected for the Analytics Innovation Centre to develop prototypes and work with the executive committee on digital transformation strategy.

• Defined requirements and built initial prototypes for Veracity, the company’s data platform connecting business consumers and suppliers.
• Selected Databricks as a data-platform vendor, evaluating the platform with its leadership team and translating business and analytical requirements into a scalable data-platform roadmap.

Led the West Coast analytics group, combining technical leadership, business development, and client delivery for major utilities and grid operators.

• Estimated peak savings across 40,000+ utility customers. Analyzed years of hourly interval data on Hadoop to support multimillion-dollar program cost recovery before the California Public Utilities Commission; filed and defended the estimates under adversarial review.
• Built a short-term renewable generation forecasting model operating at five-minute intervals to reduce grid reliance on fast-response reserves.
• Developed a discrete choice model for light bulb market shares to estimate counterfactual baselines for residential lighting programs and net-to-gross ratios.
• Led EV adopter identification for CenterPoint Energy, combining early adopter characteristics with travel survey data to target geographic concentrations.
• Built a proof-of-concept load forecasting algorithm for a major US utility’s demand response program, demonstrating the viability of daily short-term forecasts from AMI data across 40,000 customers.
• Developed load profiles and forecast errors for photovoltaic and electric vehicles for the California Independent System Operator, assessing the benefits of additional visibility into distributed energy resources.
• Managed a propensity-to-act model, combining customer attributes, consumption, and program history to estimate participation likelihood and create a new service offering.
• Piloted Apache Spark and Hadoop for large-scale energy data problems. Supported business development through direct sales and proposal strategy.

Built simulation software and estimated behavioral models for regional and statewide transportation planning projects across the US, with several supporting successful federal New Starts funding applications.

• Developed one of the first discrete choice trip distribution models for the Salt Lake City regional travel demand model, replacing the gravity-based approach and supporting a successful New Starts application.
• Led programming for the Ohio statewide travel demand model, using distributed computing to simulate long-distance travel for over 20 million people.

Supported development of the Chicago region’s 20-year long-range transportation plan through database design, GIS visualization, and travel demand model interpretation.