Staff MLOps Engineer | Ingénieur·e MLOps Staff

Job summary

The MLOps Engineer role is centered on building and scaling infrastructure for extensive 2D and 3D media datasets while ensuring the automation and efficiency of data and training pipelines. Key responsibilities include designing CI/CD and Continuous Training pipelines, managing data lifecycle strategies, and deploying systems for monitoring model performance and data drift in production environments.

Qualifications

• 5+ years of experience as an MLOps Engineer in a fast-paced environment in applied machine learning.
• Fluency with Python, Git, and the Unix shell.
• Deep familiarity with Docker, Kubernetes, and workflow orchestrators (e.g., Airflow, Prefect, or Kubeflow).
• Familiarity with collaborative tools such as Jira/Confluence, Slack, and a Git server.
• Strong mathematical background preferred for understanding resource demands of 3D data transformations.

Responsibilities

• Work with partner ML and Annotation engineers and TPMs to specify infrastructure and training requirements.
• Design and maintain robust CI/CD and Continuous Training pipelines for complex multimodal models.
• Implement versioning and storage strategies for massive 2D/3D datasets to ensure reproducibility and high-throughput access.
• Deploy and manage systems for monitoring model performance and data drift in production environments.

Skills

Degrees

Travel

Relocation