Staff MLOps Engineer

NBCUniversal•New York, NY

About The Position

We are seeking a Staff MLOps Engineer with experience building and scaling infrastructure for large 2D and 3D media datasets. You will be responsible for the "backbone" of our machine learning lifecycle, ensuring that our data pipelines are automated, reproducible, and performant at scale.

Requirements

Master's degree in Computer Science, Engineering, Mathematics, or a related field
Minimum of 5+ years of relevant industry experience, ideally within a fast-paced, high-growth tech environment.
Proven experience as an MLOps Engineer in a fast-paced environment in applied machine learning.
Prior experience in industries with complex multi-disciplinary teams such as robotics, smart grids, precision agriculture, game development, or aerospace.
Core Tools: Fluency with Python, Git, and the Unix shell.
Containerization & Orchestration: Deep familiarity with Docker, Kubernetes, and workflow orchestrators (e.g., Airflow, Prefect, or Kubeflow)
Ecosystem: Familiarity with collaborative tools such as Jira/Confluence, Slack and a Git server.
Conscientiousness: High attention to detail regarding system reliability and data security.
Systems Thinking: Ability to translate abstract ML requirements into concrete, scalable cloud or on-prem infrastructure

Nice To Haves

Strong Mathematical Background: Preferred for understanding the resource demands of 3D data transformations.

Responsibilities

Cross-Functional Coordination: Work with partner ML and Annotation engineers and TPMs to spec out infrastructure and training requirements.
Pipeline Automation: Design and maintain robust CI/CD and CT (Continuous Training) pipelines for complex multimodal models.
Data Lifecycle Management: Implement versioning and storage strategies for massive 2D/3D datasets to ensure reproducibility and high-throughput access.
Monitoring & Observability: Deploy and manage systems for monitoring model performance and data drift in production environments.