DoorDash Senior Machine Learning Scientist advancing causal models and production systems for DoorDash's Market Optimization team. Develops response curves and policies to optimize Dasher supply incentives.
Responsibilities
We are looking for a hands-on technical leader to advance the causal models and production systems behind Mobilization. You will help answer where and when to offer incentives, how much to offer, and how long they should run. You will estimate the additional supply these decisions create and translate those estimates into dependable policies with Engineering, Product, Analytics, and Operations.
You are excited about this opportunity because you will
Build and calibrate response curves that estimate how incentive amount, timing, duration, and location affect incremental Dasher hours and marketplace outcomes.
Connect causal estimates and supply forecasts to constrained optimization, balancing delivery outcomes, incentive cost, and uncertainty across markets and time periods.
Design learning experiments and evaluation plans with Analytics. Account for marketplace spillovers, supply shifting between nearby areas or time periods, and interactions with other incentives.
Develop policies that direct spending toward periods when additional supply creates the most marketplace value, accounting for diminishing returns and non-linear delivery outcomes.
Own models through production deployment, calibration, monitoring, and incident diagnosis. Build diagnostics that explain policy changes and surface model or data regressions.
We are excited about you because you have
5+ years of experience developing causal inference or causal ML systems, with a record of using the results to change production decisions.
A record of leading ambiguous projects across team boundaries, making technical tradeoffs explicit, and developing other engineers.
Strong statistical judgment about randomized experiments and observational estimation, including confounding, treatment overlap, uncertainty, and the limits of causal claims.
Experience with treatment effect estimation or dose-response modeling and relevant methods such as doubly robust estimation, double machine learning, or off-policy evaluation. You choose methods based on the decision and available evidence.
Qualification
Required
5+ years of experience developing causal inference or causal ML systems
Record of leading ambiguous projects across team boundaries
Strong statistical judgment about randomized experiments and observational estimation
Experience with treatment effect estimation or dose-response modeling and relevant methods
Strong software and ML engineering skills
Experience connecting models to constrained optimization, pricing, or resource allocation
Ability to explain modeling assumptions and incentive recommendations