Lyft is seeking to improve ad relevance, targeting, optimization, and measurement on the Lyft Ads platform to drive meaningful revenue growth and enhance marketplace efficiency.
Requirements
- Master’s or PhD in Machine Learning, Computer Science, Optimization, Statistics, Engineering, Applied Mathematics, or a related quantitative field; or equivalent high-impact industry experience.
- 5+ years of applied science or machine learning experience, with a track record of deploying production models that drive measurable business outcomes.
- Deep expertise in: Ranking and relevance modeling, CTR/CVR prediction, calibration, and uncertainty modeling, Optimization and pacing algorithms, Auction dynamics or marketplace delivery systems, Causal inference methods for ads measurement
- Strong proficiency in Python, ML frameworks (PyTorch, TensorFlow, JAX, scikit-learn), and distributed data systems (Spark, Snowflake, Databricks).
- Proven experience building large-scale, production-ready ML systems, including model servers, training pipelines, monitoring/alerting, and real-time inference services.
- Ability to define and execute offline and online evaluation strategies, including experiment design, counterfactual analysis, and diagnostics for model/system failures.
- Strong technical leadership skills — able to align partners, influence technical architecture, challenge assumptions, and guide cross-team modeling decisions.
Responsibilities
- Lead multiple high-impact Machine Learning and AI initiatives across the Lyft Ads platform — including relevance, targeting, bidding, pacing, delivery optimization, conversion prediction, and measurement systems.
- Define the modeling strategy, technical roadmap, and success metrics for ML components that power ad-serving and advertiser performance, ensuring alignment with business and revenue goals.
- Own complex, open-ended problem spaces, breaking down ambiguous advertiser, marketplace, and system constraints into well-structured modeling approaches and scientific requirements.
- Design, develop, and deploy advanced machine learning, optimization, and decisioning algorithms for large-scale real-time and batch systems, balancing scientific rigor with practical engineering constraints (latency, throughput, cost, reliability).
- Partner deeply with Ads Engineering, Infra, and Product to architect production-grade ML systems — including feature stores, training pipelines, online scoring services, monitoring, A/B frameworks, and model governance processes.
- Establish robust evaluation frameworks, defining offline metrics, calibration checks, counterfactual methods, experiment designs, and long-term measurement strategies to ensure model correctness and system stability.
- Diagnose systemic issues (drift, feedback loops, cold start, pacing imbalance, auction inefficiencies) and lead cross-functional efforts to improve model performance, user experience, and advertiser ROI.
Other
- Master’s or PhD in a related quantitative field
- 5+ years of applied science or machine learning experience
- Excellent communication skills, with the ability to articulate complex modeling concepts, system trade-offs, and scientific reasoning to both technical and business stakeholders.
- Strong technical leadership skills — able to align partners, influence technical architecture, challenge assumptions, and guide cross-team modeling decisions.
- Ability to work in-office on a hybrid schedule — Team Members will be expected to work in the office 3 days per week on Mondays, Wednesdays, and Thursdays.
