Head of Data Science: Embeddings & Measurement Leader

Jobtome; Village of Larchmont (NY); On-site; USD 263,000 - 394,000; Full time

Job summary

Spectrum is seeking a Head of Data Science to lead the development of a transformer-based behavioral embedding and related audience and measurement products. You will own model development end-to-end, govern evaluation, and partner with product and sales to scope pilots and convert evaluations into contracts.

You will build and lead a team of data scientists and ML engineers, set hiring priorities, and drive a culture of rigorous experimentation with a bias toward production deployment and

Qualifications

• Experience leading applied data science teams that build data, audience, identity, or measurement products.
• Experience with predictive modeling on consumer-level data at scale.
• Strong understanding of ML lifecycle from feature engineering to deployment.

Responsibilities

• Build the predictive models behind the platform's audience and data products.
• Lead the roadmap for the embedding model and its evaluation standards.
• Define the measurement methodology and ensure privacy-preserving analytics.

Skills

Predictive modeling at scale

Transformer-based models

Python

SQL

MLOps

Privacy-preserving analytics

Education

Bachelor's Degree in Computer Science, Statistics, Mathematics, Economics, or related quantitative field

Master's degree or PhD in Statistics, Machine Learning, Econometrics, Operations Research, or related discipline

Tools

Snowflake

LiveRamp

AWS

Databricks

Spark

Job description

Spectrum is seeking a Head of Data Science to lead the development of a transformer-based behavioral embedding and related audience and measurement products. You will own model development end-to-end, govern evaluation, and partner with product and sales to scope pilots and convert evaluations into contracts.

You will build and lead a team of data scientists and ML engineers, set hiring priorities, and drive a culture of rigorous experimentation with a bias toward production deployment and