PLEASE CLICK HERE TO SEE *ALL* OF OUR JOB OPENINGS!

Seeking a Senior Data Scientist to join a Data Science team and take ownership of model performance across core products. This role sits at the intersection of applied modeling, production systems, and real-world media performance.

You will build, configure, and improve the models that power the company’s products—from audience scoring and lookalike modeling to bid optimization and causal measurement—while owning their performance once deployed. That means diagnosing drift, running lift analyses, tuning configurations for client-specific challenges, and feeding field learnings back into the modeling cycle. Contextual targeting and NLP-based modeling will be a particular area of focus and growth.

This is not a siloed research role. You’ll work closely with engineers, product managers, and client teams, and you’ll need to be as comfortable investigating a model’s real-world performance as you are prototyping the next iteration. The ideal candidate writes production-grade code, thinks in experimentation frameworks, and understands how ad campaigns actually work.

Key Responsibilities

Modeling, Experimentation & Production Ownership

• Build, configure, and improve models across the company’s core products and own their end-to- end performance in production
• Design and execute experimentation frameworks and conduct model diagnostics using real- world campaign data
• Develop and maintain data pipelines, feature engineering workflows, and model configurations in Python, PySpark, and MLflow on Databricks
• Prototype and build agentic AI systems that extend product capabilities

Cross-Functional Collaboration & Product Impact

• Serve as the data science expert across client and product teams—translating model outputs and campaign KPIs into actionable recommendationsSurface edge cases, performance drift, and feature requests from the field to inform core DS roadmap priorities
• Contribute to research cycles, internal tooling, and DS best practices

Required Qualifications

• 5+ years in data science or machine learning roles. Master’s degree preferred.
• Experience with NLP, contextual modeling, or embedding-based approaches for modelling
• Ability to write production-grade, testable code beyond notebooks
• Strong proficiency in Python, large dataset processing (PySpark and SQL), modelling libraries (PyTorch, sklearn) in a Databricks or AWS.
• Excellent communication skills; ability to translate technical findings for non-technical stakeholders

Nice to Have

• Strong grasp of statistical analysis, causal inference and experimentation design
• Working familiarity with MLflow (experiment tracking, model registry)
• Experience and familiarity with adtech and digital advertising
• Experience building agentic AI workflows, LLM-based systems, or tool-use orchestration
• Experience with CI/CD workflows (GitHub Actions) and software packaging best practices

Tech Stack

• ML Frameworks: PyTorch, Sklearn, Ray (Train, Tune, Datasets), PySpark ML
• Data Platform: Databricks (Delta Lake, Unity Catalog), Snowflake, AWS (S3, EC2)
• MLOps: MLflow (experiment tracking, model registry), GitHub Actions
• Languages: Python, SQL, JavaScript/TypeScript

Job-3547684

*LI-MG1

#LI-Remote