A Tenth Revolution Group Company

Ref: 15092023

MLOps Engineer


  • Negotiable
  • Engineer Role
  • Skills: Pytorch, OpenCV, Tensorflow, Kubeflow, MLflow, TFX, Airflow
  • Level: Mid-level

Job description

MLOps Engineer



We're at the forefront of an exciting project, leveraging technology to bring groundbreaking solutions to life by harnessing the power of technology, Transforming the traffic system

  • Education: BS, MS, or PhD in Machine Learning, Computer Science, Electrical Engineering, or a related field.
  • Experience: Minimum 3 years in MLOps, specializing in computer vision applications, including ML model development, deployment, and monitoring.
  • Skills: Proficiency in Python. Hands-on experience with ML pipeline tools such as Kubeflow, MLflow, TFX, Airflow, and more. Familiarity with common libraries like Pytorch, OpenCV, Tensorflow, and others.
  • Data Handling: Proficiency in managing large-scale datasets for computer vision tasks. Demonstrated expertise in developing auto-annotation tools for visual data.
  • Collaboration: Experience in working with external and internal data annotation services.

Nice to Have:
  • Knowledge of active learning, semi-supervised learning, and similar approaches for visual data analysis.
  • Demonstrated experience in MLOps best practices and automation.
  • General understanding of DL model development and deployment on embedded platforms and cloud solutions.
  • Experience with multi-task and semi-supervised DL model training on video data.
  • Proven track record in industry experience or publications in relevant conferences.
  • Proficiency in additional languages is a plus.

  • Design and develop data scanners and auto-annotation engines for optimizing computer vision datasets.
  • Collaborate with data annotation teams to enhance dataset production.
  • Create, deploy, and automate ML pipelines for efficient model training and deployment.
  • Monitor model performance and manage model and dataset versioning for operational efficiency.
  • Participate in end-to-end development, from defining problems to model design, deployment, and continuous improvement.