Principal Machine Learning Engineer, Recommendation Systems
Launch Potato
Open Source
Principal ML Engineer, Recommendation Systems
Launch Potato
full-time
Posted on:
Origin: • 🏈 Anywhere in North America
Visit company websiteJob Level
Lead
Tech Stack
Open Source
About the role
- Establish the technical vision for personalization at Launch Potato, solving the company’s most complex ML challenges and influencing strategy, architecture, and innovation across teams
- Define company-wide personalization strategy and architecture, driving alignment across all ML teams
- Solve critical technical challenges such as cold start, real-time learning, and exploration/exploitation tradeoffs
- Design and implement advanced ML solutions using cutting-edge techniques (e.g., graph neural networks, causal models, bandit algorithms)
- Create and enforce ML architecture patterns, design standards, and reusable infrastructure across teams
- Lead multi-quarter, cross-functional initiatives that redefine how personalization impacts business KPIs
- Act as technical mentor to senior ML engineers, guiding complex decision-making and scaling team capability
- Represent Launch Potato’s technical brand externally through speaking engagements, open source contributions, or publications
- Champion privacy-preserving personalization, responsible AI practices, and adaptive learning systems
Requirements
- Expertise building ML systems with deep expertise in large-scale personalization
- Recognized industry expertise through patents, publications, or significant product impact
- Proven success architecting ML platforms serving billions of predictions in production
- Demonstrated track record of 0→1 innovation in personalization or recommender systems
- Mastery across deep learning, causal inference, multi-armed bandits, and graph-based models
- 10+ years designing, developing, and deploying large-scale machine learning systems with a focus on personalization
- Strong technical leadership and mentorship experience guiding senior ML engineers
- Experience addressing cold start, real-time learning, and exploration/exploitation tradeoffs
- Experience designing systems using graph neural networks, causal models, and bandit algorithms
