Description

Design, build, and operate low-latency ML serving systems for fraud scoring, personalization, and identity resolution at scale.

This role is on-site.

Responsibilities

• Architect and maintain probabilistic identity resolution systems connecting device IDs and first-party cookies to households.
• Build ML models for audience intelligence, content affinity, genre-preference, and time-series forecasting across global markets.
• Design feature pipelines that fuse real-time streaming signals with batch-computed features for online scoring.
• Lead technical roadmap shaping, execution planning, and capability development while mentoring senior engineers.
• Evaluate new technologies including graph ML, agentic ML orchestration, and LLM-augmented pipelines for production use.

Required Skills

• 9-13 years of ML engineering experience, or 6+ years with a Ph.D., demonstrating Staff-level scope.
• Proficiency in Python, PyTorch, TensorFlow, Spark, and MLOps.
• Experience with AWS, Databricks, Snowflake, and Sagemaker.
• Hands-on experience with optimization problems like ranking, personalization, or multi-objective decisioning.
• Strong background in building production ML systems for large user populations with cross-team influence.
• Experience designing offline and online evaluation approaches with clear baselines and experiment design.
• Master's or Ph.D. in Computer Science, Statistics, Machine Learning, or equivalent industry experience.

Preferred Skills

• Experience with XGBoost, LightGBM, scikit-learn, and real-time feature serving.
• Knowledge of graph neural networks, mixture-of-experts, and agentic AI frameworks.
• Background in ad-tech ML, fraud detection, recommendation systems, or streaming platforms.

Key Skills

Education

Any Graduate