Jumio
Research Engineer – Machine Learning & Robotics
About this role
Jumio is seeking a Research Engineer to develop and scale robotic data collection systems and ML infrastructure for computer vision and identity verification products. You'll work hands-on with ROS-based robotics, build ML pipelines, and ensure high-quality datasets that drive model improvement across mobile and robotic platforms.
What you'll do
- Build and integrate ROS/ROS2 modules for robotic navigation, manipulation, and automated data collection
- Develop and maintain training datasets from robotic systems and iOS/Android applications
- Create tools to monitor data quality, model performance, and accuracy in production environments
- Mine and analyze internal database data to generate features, identify trends, and improve model development
- Implement end-to-end ML workflows including data preparation, training, testing, and deployment support
- Write clean, modular Python and C++ code for robotics and ML infrastructure
What they're looking for
- ROS/ROS2 framework
- Python and C++ programming
- Machine learning fundamentals and frameworks (PyTorch, TensorFlow, scikit-learn)
- Computer vision
- Database design and SQL queries
- Data pipeline development
- Hardware debugging and integration
- ML model evaluation and monitoring
Opens the application — the Jobs AI extension fills it for you. Set up autofill
Opens the official application on the employer’s site. No login required.
Jumio
Jumio builds computer vision and identity verification products powered by machine learning, with infrastructure spanning mobile and robotic platforms. The company is hiring Research Engineers to develop robotic data collection systems, ML pipelines, and datasets that improve identity verification models.
View all jobs at JumioLikely interview questions
- Walk us through a project where you built or integrated ROS/ROS2 modules. What was the most challenging part, and how did you debug it?
- Describe your experience implementing machine learning models end-to-end—from data preparation through deployment. Which frameworks have you used, and why?