PubMatic

Machine Learning Engineer

• Build and improve machine learning models for campaign optimization, prediction, ranking, bidding, forecasting, and calibration.
• Develop models and algorithms that improve advertiser outcomes while balancing spend delivery, cost efficiency, campaign goals, marketplace dynamics, and system constraints.
• Work on large-scale ML systems using signals from auctions, impressions, clicks, video events, conversions, users, context, inventory, campaigns, and marketplace feedback.
• Design and improve CTR, CVR, VCR, CPA, ROAS, app-install, user-value, and campaign-performance models.
• Develop bidding, pacing-aware optimization, ranking, exploration, and value-estimation approaches for performance advertising.
• Improve model calibration, online/offline evaluation, experimentation, observability, and production feedback loops.
• Reason through sparse conversions, delayed feedback, biased logs, cold-start campaigns, attribution noise, and online/offline metric mismatch.
• Partner with performance advertising signal engineers to define model-ready features, labels, attribution windows, negative examples, training datasets, and online serving requirements.
• Partner with engineering, product, analytics, and platform teams to translate model outputs into real-time decisioning systems.
• Help evolve Activate from a media buying execution platform into a performance optimization platform.
• Provide technical leadership and mentorship to engineers and applied scientists working on performance optimization problems.

Senior Principal Machine Learning Engineer – Optimization