The Trade Desk is seeking to solve the problem of providing accurate forecasting models to support planning and recommendation-related products across their advertising platform. Accurate forecasting is critical for timely feedback on user selection in audience, inventory, and other campaign setups, which is essential for delivering intelligent advertising at scale.
Requirements
- Proficient in machine learning (ML) and deep learning (DL). Hands-on development of ML and DL solutions utilizing open-source tools and cloud computing platforms
- Hands-on experience building deep learning models at large scale
- Experience in deep learning, PyTorch/ TensorFlow preferred
- Experience running heavy workloads on a distributed computing cluster (especially EMR or Databricks), leveraging technologies like Spark to work with large datasets preferred
- Proficient in Python. Strong Spark skills are a plus
Responsibilities
- Develop ML/deep learning based models end-to-end to support various forecasting needs throughout advertising campaign life cycle, including working on signal collection, feature construction, model evaluation and monitoring
- Collaborate with engineer team on signal collection pipeline and model training, deployment and monitoring processes
- Collaborate with downstream team to understand their forecasting needs and design appropriate models
- Design offline and online model success metrics and use data to drive iterative model improvement
Other
- BS/MS with 4+ years or a PhD with 2+ years of experience working in a DS or ML role that involves bringing products from ideation to production
- The ability to communicate with diverse stakeholders, making architecture recommendations, ensuring effective execution, and measuring quality of outcomes
- Familiar with forecasting domain with solid knowledge background and application experience
- Possessing a keen sense of data intuition and the ability to innovate in the field of model development
- Knowledge in privacy safe measurement and data storage is a plus
