Principal Machine Learning Engineer, Ads Personalization

Paramount is on a mission to unleash the power of content and is seeking a Principal Machine Learning Engineer to lead the Ads Pod, focusing on architecting their proprietary ad system. The role involves leading a team to develop algorithms for ad selection and optimization across their streaming platforms, balancing user experience with business goals.

Responsibilities:

• Lead the Ad Pod: Drive the technical roadmap for all Pluto-specific ML use cases, acting as the bridge between AMLG's core infrastructure and Pluto’s product goals
• Personalized Channel Discovery: Architect retrieval and ranking systems to recommend the right linear channels to users in real-time
• Dynamic Scheduling & Creation: Develop models that optimize channel scheduling and inform the creation of "Pop-up" or algorithmic channels based on viewer trends
• Multi-Objective Optimization: Design loss functions and reward systems that weigh user commitment (watch time) against business health (ad impressions and revenue)
• EPG Personalization: Lead the vision for personalizing the Electronic Programming Guide, ensuring the most relevant content is surfaced immediately upon app launch
• Cross-Functional Leadership: Partner with RevOps and Product to translate business constraints (e.g., ad delivery guarantees) into technical ML requirements

Requirements:

• 6-8+ years of experience in machine learning engineering, with a significant focus on Ad-Tech, Auction Dynamics, or Recommender Systems
• Deep mastery of Multi-Objective Optimization and constrained optimization (balancing competing KPIs like revenue vs. UX)
• Proven experience building and deploying ML models in high-throughput, ultra-low latency environments (under 50ms)
• Skilled in Python, PyTorch/TensorFlow, and BigQuery; experience with high-scale serving layers
• Demonstrated ability to own a major technical domain and drive strategy across multiple organizations
• Experience building or scaling in-house ad systems or DSP/SSP components
• Knowledge of Reinforcement Learning (RL) for sequential ad-podding and frequency capping
• Knowledge of Causal Inference to measure the incremental boost ad-personalization on long-term subscriber churn
• Experience in both FAST (Linear) and VOD advertising ecosystems