Transforming AI ambition into measurable product impact.

Research, Measurement & strategy

Defining how Agentic Search is measured and evaluated

Contributing to development of next-generation of Agentic experiences on Google Search

Leading Data Science strategy for emerging Agentic Search experiences ↗, building the measurement and evaluation systems needed to assess how effectively Search helps people navigate complex, multi-step tasks and move from discovery to action.

Developed an integrated framework spanning experimentation, behavioral signals, user satisfaction, and human-and-AI evaluation to define product quality, quantify user value, and identify opportunities for sustainable growth. Partnered with Product, Engineering, Research, and senior leaders to guide product priorities, evaluate experience quality, and enable evidence-based decisions at scale.

Metrics Established multidimensional quality frameworks spanning helpfulness, relevance, factuality, and groundedness eval metrics, providing teams with a more complete view of user value.

Research Advanced research into how users perceive value in complex AI-assisted journeys and how successful task outcomes can be evaluated consistently.

Experimentation & Metrics Research

Reframing success metrics for AI-Native Search

Building a decision framework for next-generation AI Search

A fundamental shift from traditional search results to an AI-orchestrated experience made existing Search engagement and quality metrics difficult to interpret. Changes in user behavior could reflect genuine product quality or simply novelty, change aversion from habits formed around the previous experience. We led the development of a new measurement framework that enabled fair comparisons across fundamentally different product experiences. The approach combined meaningful interactions, satisfaction signals, and human quality evaluations into decision-ready launch criteria. Partnering across Product, Engineering, UX Research, and evaluation teams, we established validation methods and metric priorities that helped leadership distinguish durable user value from transitional behavior and make more confident launch and investment decisions.

Key Impact Reframed measurement for next-generation AI Search by unifying engagement, satisfaction, and human-quality signals to distinguish durable user value from adaptation effects enabling confident launch and investment decisions.

AI Mode & Product Strategy

AI-Powered Finance: Reimagining Google Finance with AI

Advancing Finance AI respose quality through meta-prompting

Led Google’s effort to transform Google Finance ↗ into a personalized daily destination for investors—combining high-quality data, live news feeds, and advanced AI-powered research tools. Directed the data science roadmap and evaluation architecture for the Research Panel, the integration of AI Mode into current Google Finance experience. This executive-level product priority, designed for active investors, has now launched globally.

Led the development of a novel meta-prompting framework that transformed Finance AI responses into concise, contextual, and data-driven financial analysis — shaping both the product experience and its quality-evaluation strategy.

Key Impact Our approach resulted in a more sophisticated production prompt that transformed Finance Hub AI’s persona into an expert financial analyst.