Principal Data Scientist, Recommendations

• Build recommendation systems that identify likely drivers of underperformance, recommend creative improvements, prioritize edits, and support testing
• Develop creative intelligence systems connecting visual, text, audio, structural, platform, and performance data
• Turn multimodal signals into diagnostics, scores, recommendations, and decision-support tools
• Work with video, image, text, audio, metadata, and KPI data using computer vision, embeddings, LLMs, classification, clustering, and multimodal reasoning
• Design experiments, quantify uncertainty, and separate signal from noise using causal inference, hypothesis testing, regression, matched cohorts, and treatment/control analysis
• Write Python and SQL for data products, APIs, model pipelines, evaluation frameworks, and internal tools
• Partner with Engineering, Data Engineering, and DevOps on reliable, observable, scalable, and maintainable systems
• Improve training pipelines, feature stores, model serving, evaluation harnesses, and deployment workflows across AWS and GCP / Vertex AI
• Serve as a technical leader and explain complex AI, machine learning, and measurement concepts to cross-functional teams
• Help the organization adopt better tools, automation, evaluation methods, and AI-assisted development practices

• PhD in Computer Science strongly preferred; PhD in statistics, mathematics, natural science, engineering, operations research / management science, or economics valued with strong CS and engineering experience; exceptional MS candidates considered with substantial applied AI / ML experience
• 10+ years in applied data science, machine learning, or AI product development, or 5+ years post-PhD in a highly technical applied role
• Production-quality Python
• Experience designing data systems, reasoning about architecture, and partnering with engineers on deployed data products
• Ability to extract information from large data tables and manipulate data using SQL queries and scripting code
• Hands-on experience with Databricks, BigQuery, Flink, or similar large-scale data platforms
• Experience with model training, evaluation, deployment, monitoring, retraining, feature pipelines, experiment tracking, model registries, and ML observability
• Strong command of supervised and unsupervised learning, model evaluation, feature engineering, embeddings, retrieval, ranking, clustering, classification, recommendation systems, and statistical validation
• Practical experience with computer vision, video/image understanding, LLMs, foundation-model workflows, or other applied GenAI systems
• Strong command of probability, statistics, optimization, experimental design, causal inference, hypothesis testing, confidence intervals, and treatment/control design
• Experience applying models to real business outcomes
• Strong written and spoken English skills