Software Engineer, Conversion Measurement

Responsibilities

• Design and build reliable conversion-event collection and processing systems across pixels, SDKs, server-side APIs, app events, offline uploads, and advertiser integrations.
• Build attribution and measurement infrastructure for observed and modeled conversions, attribution windows, deduplication, delayed events, and signal-loss mitigation.
• Develop privacy-preserving identity, matching, aggregation, and reporting systems that provide useful measurement under strict privacy constraints.
• Create data-quality systems that detect missing, duplicated, malformed, out-of-order, fraudulent, or misconfigured conversion signals.
• Build scalable advertiser reporting and diagnostics for conversions, cost per action, return on ad spend, funnel performance, and measurement health.
• Partner with Ads ML to deliver trustworthy conversion labels and feedback loops for ranking, bidding, targeting, and budget optimization.
• Develop experimentation and incremental capabilities, including holdouts, lift studies, and causal measurement foundations.
• Define the technical strategy and roadmap for conversion measurement across OpenAI’s ads delivery stack.
• Operate systems with high engineering rigor through testing, observability, privacy reviews, incident response, and operational best practices.

Requirements

• 10+ years experience building and operating large-scale distributed data or backend systems, ideally in ads measurement, attribution, analytics, marketplaces, experimentation, or adjacent domains.
• Understand conversion-event pipelines, attribution, deduplication, identity and matching, reporting, or downstream optimization signals.
• Experience designing privacy-safe measurement systems, including aggregation, consent handling, modeled conversions, clean-room patterns, or privacy-preserving APIs.
• Can reason about data correctness, delayed and out-of-order events, schema evolution, reconciliation, and internal-versus-external metric discrepancies.
• Have built high-throughput batch or streaming systems and can make sound tradeoffs across latency, reliability, cost, and maintainability.
• Comfortable partnering with ML and data science teams on labels, model inputs, experimentation, and causal measurement.
• Think holistically across architecture, product semantics, data quality, observability, privacy, and advertiser trust.
• Can define technical direction in an ambiguous 0→1 environment and independently drive complex work across teams.
• Communicate clearly and make technical decisions grounded in user value, system health, measurement credibility, and long-term product direction.

Nice-to-Haves

• Experience in ads measurement, attribution, analytics, marketplaces, experimentation, or adjacent domains.
• Background building privacy-safe measurement systems.