- Develop and improve ads ranking models, including prediction objectives, feature interactions, user-history modeling, and calibration.
- Take end to end ownership of machine learning systems - from data pipelines, feature engineering, training-data construction and model evaluation, model training, as well as integration into our production systems.
- Evaluate and apply advances in deep learning and recommendation modeling to improve ads ranking within production latency, reliability, and cost constraints.
- Collaborate with ML platform and product engineers to build scalable and efficient machine learning systems in the production environment.
- Partner with product, data science, and engineering teams to define ranking objectives, design A/B experiments, and measure improvements in advertiser performance, revenue, and user relevance.
- Identify new opportunities to apply machine learning to different parts of the Ads product to drive value for our users and advertisers.