Senior Data Scientist

Who you are

• MS or PhD in Operations Research, Mathematics, Statistics, Economics, Computer Science, Engineering or a related field, with 2+ years of experience in delivery, logistics, marketplace or pricing problems (or 8+ years of industry experience in data science)
• Strong background in at least two of the following: pricing/revenue optimization, mathematical optimization (e.g., MIP, assignment and routing problems, heuristics), time-series and demand forecasting, experimentation and causal inference
• Hands-on experience designing experiments in two-sided or operational marketplaces where standard A/B tests break down (e.g., switchback or cluster-randomized designs, variance reduction)
• A proven ability to translate ambiguous business problems into clear technical objectives and deliver concrete, measurable results
• Expertise in machine learning and statistics
• Proficiency in Python, SQL and source control; experience shipping models to production with engineering teams
• Experience in on-demand delivery, ride-hailing or gig-economy marketplaces
• Familiarity with optimization solvers (e.g., OR-Tools, Gurobi) and forecasting libraries
• Experience with ETA prediction or driver incentive and pay design

What the job involves

• Optimize delivery fees: Design, build and iterate on algorithmic delivery fee strategies (e.g., dynamic or contextual pricing by location, time, basket and capacity) that balance conversion, customer experience and profitability
• Solve the driver dispatch problem: Develop optimization and ML approaches to order batching, assignment and routing that lower overall delivery cost while keeping speed and reliability high
• Forecast demand: Build accurate, well-calibrated demand forecasts at the right granularity (location × time) to inform driver supply planning, scheduling, incentives and pricing decisions
• Run rigorous experiments: Design and analyze high-quality tests, including A/B, switchback and geo/market-level designs, that account for marketplace interference, and use causal inference to measure true impact on customer and business metrics
• Own solutions end to end: frame the problem, build features and models, deploy with engineering partners, then monitor, explain and improve performance in production
• Partner with product, engineering, operations and finance to turn ambiguous business questions into well-defined technical objectives and shipped solutions
• Deeply understand our customers, delivery partners and data to find new opportunities for ML, optimization and causal inference