Principal Product Manager, AI Evaluation and Data Policy

The role

You will own Person Data outcomes end to end, including strategy, roadmap and priorities for contact accuracy, coverage, freshness and compliance. You will build the AI evaluation discipline for the Core Data team and lead the move from rules to agents.

What you will do

• Own Person Data strategy, roadmap and monthly priorities for accuracy, coverage, freshness and compliance
• Build and run the AI evaluation engine, including definitions of correct, golden sets, LLM-as-judge gates and confidence thresholds
• Lead the pivot from rules to agents, writing policy clauses and retiring legacy selection logic once data supports it
• Decide where AI fits in the pipeline and where classical tooling is better, weighing cost, latency, determinism and auditability
• Lead entity resolution work on canonical identity, match confidence and the trade-off between false merges and missed matches
• Own the privacy roadmap for person data, including suppression, opt-out and notification, with Legal and Privacy
• Prototype adjudication agents and evaluations in code alongside the AI engineer and data science team
• Act as the hub across Person Data engineering, Match, Research, Data Science, Web Data acquisition and Privacy

Required qualifications

• 8+ years of product management experience
• Meaningful time owning data products or data platforms at scale
• Hands-on experience with data pipelines and how records move from ingestion through processing to a served dataset
• Direct experience with entity resolution, record linkage or matching systems, including canonical IDs, match confidence, survivorship rules and precision/recall trade-offs
• Demonstrated, current use of AI to improve data systems in production, and a way to describe how accuracy was kept from degrading
• Experience designing evaluations for model output, including golden sets, precision and recall, LLM-as-judge, human-in-the-loop routing and production drift monitoring

Nice to have

• Experience with incremental or streaming data pipelines, rather than batch rebuilds
• Experience with contact, person or identity data, or with B2B company data
• Familiarity with privacy regulation for published personal data (GDPR, CCPA, DNC and international notification regimes)
• Experience with Claude or comparable frontier models, agent frameworks and eval harnesses

Benefits

• Comprehensive benefits
• Holistic mind, body and lifestyle programs for overall well-being

Before you apply

• ZoomInfo may use a software-based assessment as part of recruitment; the posting links to a bias audit
• Massachusetts applicants: lie detector tests are not administered to applicants

About the company

ZoomInfo (NASDAQ: GTM) is an all-in-one AI go-to-market platform that helps sales, marketing and customer success teams identify, engage and convert buyers. It describes its B2B data as covering more than 100 million companies and 500 million contacts.