Senior Machine Learning Engineer, Ads Response Prediction

What you'll do

• Develop and execute research for pCTR and conversion prediction models to improve accuracy across ad surfaces.
• Implement debiasing techniques like Mixed Negative Sampling and Inverse Propensity Weighting to address training data biases.
• Build multi-task learning architectures using Mixture-of-Experts, Transformer layers, and LoRA adapters for domain fine-tuning.
• Develop sequence modeling initiatives including generative retrieval systems and Semantic ID representation learning.
• Contribute to the development of foundation models using autoregressive user behavior prediction.
• Translate ambiguous business problems into well-defined machine learning research directions with clear evaluation criteria.
• Present technical findings internally through design reviews, paper sharing, and experiment retrospectives.

What we're looking for

• Master's or PhD in machine learning, statistics, computer science, information retrieval, or a related quantitative field; or equivalent experience.
• 3+ years of combined academic and industry experience applying ML to ranking, recommendation, or prediction problems at scale.
• Deep understanding of CTR/conversion prediction modeling including architectures like Deep & Wide, DeepFM, DCN, and multi-task learning.
• Strong foundation in causal inference, counterfactual reasoning, and training data bias mitigation.
• Proficiency in Python, deep learning frameworks (PyTorch, TensorFlow, JAX), and data manipulation tools (SQL, Spark, Pandas).
• Track record of formulating ambiguous problems into well-scoped ML research directions and delivering results through rigorous experimentation.
• Strong written and verbal communication skills to explain complex modeling decisions to cross-functional stakeholders.
• Experience in ads ranking, autoregressive sequence models, or publication records in top-tier venues (preferred).