Sr. Applied Scientist

About the role

Build reinforcement learning and contextual bandit models that drive real-time ad selection, ranking, and bidding for a programmatic advertising platform. Design offline and online experiments, including off-policy evaluation, and deploy models that run reliably at high throughput and low latency. Requires 3-5 years of machine learning experience, strong Python and PyTorch or TensorFlow skills, and a background in reinforcement learning or sequential decision-making.