Improving the Uber user experience and overall marketplace performance by solving key business problems such as designing incentives structures for drivers, measuring causal impacts of promotions, and creating ML-driven optimizations for marketplace growth
Requirements
- Experience in ML frameworks (e.g. Tensorflow, Pytorch, or JAX) and complex data pipelines
- Programming languages such as Python, Spark SQL, Presto, Go, Java
- Strong problem-solving skills, with expertise in ML methodologies
- Experience in applying ML, statistics, or optimization techniques to solve large-scale real-world problems (e.g. ads tech, recommender systems)
- Experience in designing and crafting scalable, reliable, maintainable and reusable ML solutions using deep-learning techniques and statistical methods
- Detail-oriented, ownership and truth-seeking mindset
- Values and produces analytic evidence and insight, as well as applying them to improve technical solutions
Responsibilities
- Develop and optimize ML models to enhance key marketplace signals (e.g., ETA predictions, supply availability metrics, demand forecasts)
- Collaborate with cross-functional teams (Pricing, Matching, Driver Incentives, etc.) to ensure marketplace signals are effectively utilized
- Improve operational efficiency by building a centralized, scalable system for marketplace signals that serves multiple use cases
- Leverage cutting-edge ML techniques (deep learning, probabilistic modeling, reinforcement learning, etc.) to continuously refine marketplace signals
Other
- B.S. in Statistics, Mathematics, Computer Science, or Machine Learning
- 2 years of experience in software engineering with an emphasis on data-driven methodologies, deep learning, and online experimentation
- Master's degree in Computer Science, Engineering, Mathematics or related field (preferred)
- 3+ years of experience in software engineering specializing in applied ML methods (preferred)
- Ability to work in a cross-functional and/or cross-business projects, partnering with Product, Scientists, and cross-org leads to shape the team's strategies
