Head of AI Data Science, Intelligence Ventures

Location: New York

This role requires the ability to work lawfully in the U.S. without employment-based immigration sponsorship, now or in the future.

JOB SUMMARY

The Head of AI Data Science serves as the head of AI research and leader of data science operations for a new behavioral intelligence platform initiative within Charter Communications. This executive owns the design, training, validation, and real-world application of the platform's proprietary transformer-based behavioral model - the engine that converts household-scale network signals into the embeddings and predictive features that power the platform's intelligence products.

The Head of AI Data Science leads a team of data scientists and ML researchers, partners closely with the Head of Technology on infrastructure, and works in direct collaboration with external development partners during the initial build phase. This role sits on the platform leadership team and reports directly to the Head of Intelligence Ventures.

HOW THE HEAD OF AI DATA SCIENCE MAKES AN IMPACT

• Direct the research, design, and training of the platform's proprietary transformer-based behavioral embedding model - a multi-entity architecture that encodes household behavior across multiple signal sources into dense, privacy-safe vector representations. Own the full model development lifecycle from architecture decisions and training methodology through validation, deployment, and ongoing iteration as new signal sources and use cases are introduced.
• Lead the design and build of the platform's Feature Store - translating embedding representations into interpretable, actionable behavioral signals including purchase propensities, category interest intensities, lifestyle affinities, and behavior velocity signals. Oversee the outcome anchoring methodology that trains predictive models against external third-party datasets to produce validated, commercially relevant intelligence outputs across target verticals.
• Partner with the Head of Technology and external development partners to ensure the AI/ML architecture is production-grade, built for household-scale throughput, and integrated cleanly into the platform's cloud-native infrastructure. Establish model evaluation frameworks, quality benchmarks, and MLOps practices that enforce a strong bias toward production-deployed, commercially validated outputs - not just research-quality results.
• Serve as the platform's primary AI research voice in external partner conversations - including technical engagements with cloud AI platforms, frontier model teams, and enterprise data partners - articulating the platform's embedding architecture, signal differentiation, and model enrichment value proposition to sophisticated technical counterparts. Contribute to the development of packaged intelligence products such as behavioral demand indices, persona clusters, and predictive propensity scores.
• Establish the platform's responsible AI framework - including bias testing protocols for behavioral embeddings, model documentation standards, and privacy-preserving ML techniques - ensuring all intelligence products meet ethical and regulatory standards for consumer behavioral data.
• Build and lead a team of data scientists and ML researchers capable of competing with talent from the world's leading AI research and applied ML organizations. Establish the team's research agenda, hiring priorities, and culture of rigorous experimentation - maintaining a clear bias toward applied, production-oriented work while preserving the intellectual ambition required to stay ahead of a rapidly evolving AI landscape.

QUALIFICATIONS

REQUIRED QUALIFICATIONS

• Deep expertise in transformer-based sequence modeling and its application to behavioral or interaction data at consumer scale - including architecture design, training methodology, fine-tuning, and embedding quality evaluation
• Proven track record developing and deploying household- or user-level embedding models applied to real-world use cases in media, marketing, commerce, and/or customer intelligence - not just research environments. Demonstrated understanding of the unique characteristics of behavioral sequence data: sparsity, temporal dynamics, multi-entity structure, and the signal differences between behavioral intent and explicit interaction
• Strong command of the full data science lifecycle in production settings - from exploratory data analysis and feature engineering through model training, validation, deployment, monitoring, and iteration - at large dataset scale (billions, even trillions of records)
• Hands-on proficiency with Python, PyTorch or Tensor Flow, and distributed ML training frameworks; experience running ML workloads on cloud platforms (AWS Sage Maker, Snowflake Cortex, Databricks, or equivalent)
• Experience designing and operationalizing feature stores and predictive modeling pipelines that serve downstream intelligence products, audiences, or decision systems in production environments
• Ability to communicate complex AI/ML concepts…