Staff Machine Learning Engineer - Ads Monetization

Reddit is home to over 100,000 active communities and approximately 130 million daily active unique visitors.

As a Staff Machine Learning Engineer on the Ads Forecasting team, you will take end-to-end technical ownership of the machine learning capabilities and production pipelines that allow advertisers to predict campaign performance, simulate spending scenarios, and optimize their investments. In this hands-on, high-impact role, you will connect inventory supply, audience demand, and response prediction—building robust models that account for auction dynamics, budget shifts, cold start, delayed feedback, and marketplace competition.

Key Responsibilities

• End-to-End ML Ownership: Own forecasting and scenario modeling capabilities from problem formulation and technical architecture to evaluation, production deployment, and continuous iteration.
• Scenario & Outcome Modeling: Build and refine machine learning models that estimate how campaign outcomes change across alternative budgets, bids, audience segments, creative assets, and optimization targets.
• Inventory & Response Integration: Synthesize signals across available advertising inventory, user behavior, audience activity, and response models to deliver accurate outcome predictions.
• Generalization & Sparsity Resolution: Engineer techniques to handle challenging modeling conditions—such as cold start, sparse audience segments, seasonality, distribution shift, and delayed feedback.
• Rigorous Model Evaluation: Measure model accuracy, calibration, and prediction uncertainty across advertiser cohorts using temporal validation and off-policy evaluation techniques.
• Production Systems & Mentorship: Operate training, inference, and real-time monitoring infrastructure for production ML models while providing technical guidance and code reviews for team members.

Requirements

Machine Learning & Predictive Modeling Depth

• Proven track record independently designing, shipping, and operating production ML systems in advertising, marketplace platforms, recommendation systems, or search.
• Strong mathematical foundations in statistical learning, probability theory, and mathematical optimization using large-scale behavioral or transactional datasets.
• Hands-on expertise in relevant domain areas, such as time-series forecasting, probabilistic modeling, response prediction, multitask learning, or transfer learning.
• Sound evaluation judgment—including segment-level analysis, temporal validation, calibration metrics, and handling uncertainty.

Software Engineering & Execution

• Strong software engineering skills with experience building and operating production ML pipelines, feature stores, and microservices.
• Demonstrated ability to translate ambiguous business and product objectives into concrete technical plans and execute them through rollout.

Preferred Qualifications

• Experience in demand/supply/inventory forecasting (time-series, hierarchical/global models, multi-horizon predictions).
• Expertise in auction and investment modeling (auction simulation/replay, win-rate prediction, bid/budget response curves).
• Familiarity with counterfactual or off-policy evaluation, sensitivity analysis, and uncertainty propagation.
• Domain depth in reach and audience estimation (population overlap, reach/frequency saturation models, sampling correction).
• Hands-on experience with probabilistic response modeling (quantile prediction, calibrated conversion estimation, or delayed-feedback modeling).

Total Rewards & Benefits

• Compensation & Equity: Base salary of $216,700 – $303,400 USD + Restricted Stock Unit (RSU) equity grants.
• Healthcare & Wellbeing: Comprehensive medical, dental, and vision insurance; family planning support; gender-affirming care; and mental health/coaching benefits.
• Retirement & Income Protection: 401(k) program with employer match, life insurance, and income replacement programs.
• Time Off & Work Culture: Flexible vacation policy, paid volunteer time off, global lifestyle reimbursements, and generous paid parental leave.