
Principal Machine Learning Engineer
Posted Jul 28

Posted Jul 28
This is a fully remote position, open to applicants in New York.
• Develop the technical strategy: Take ownership of the multi-quarter technical roadmap for personalization, which includes candidate generation, ranking, and exploration.
• Design and implement end-to-end systems.
• Create multi-stage personalization systems that encompass retrieval, deep ranking, contextual embeddings, and bandit-based exploration.
• Oversee the complete process, from feature engineering through training, serving, and monitoring.
• Enhance modeling practices in production environments.
• Employ advanced methodologies involving representation learning, multi-task learning, multi-modal embeddings, and session modeling to boost user satisfaction.
• Collaborate with Core Science, Content Understanding, ML Platform, and Product teams to ensure personalization aligns with the overarching strategy.
• Guarantee that personalization pipelines are high-throughput, dependable, and observable in GCP utilizing TensorFlow/PyTorch and big-data tools (Beam, BigQuery).
• Elevate experimentation quality and enhance the correlation between offline and online results.
• Mentor engineers and cultivate the next generation of senior ML professionals.
• Minimum of 8 years of experience in Machine Learning Engineering (MLE), applied science, or large-scale recommender/ranking systems, demonstrating a history of setting technical direction.
• Expertise in representation learning.
• Experience with multi-modal embeddings.
• Understanding of contextual bandits and session modeling.
• Strong proficiency in experimentation methodologies, A/B testing, causal reasoning, and metric design.
• Skilled in GCP, TensorFlow, and PyTorch.
• Proven ability to influence technical strategy across various teams.
• Medical
• Dental
• Vision
• 401(k) plan
• Life insurance coverage
• Disability benefits
• Tuition assistance program
• Paid Time Off (PTO)
NVIDIA
SentiLink
SentiLink
Leega
Get handpicked remote jobs straight to your inbox weekly.