Job summary
This role involves developing machine learning and computer vision models to enhance an autonomous agricultural robot for sustainable farming. The engineer will handle data from real-world farming environments, improve model reliability across varying conditions, and collaborate closely with multidisciplinary teams to deploy practical AI-powered solutions in agriculture.
Location and work setup
- Location
- Norway
- Work setup
- On-site
Responsibilities
Design, train, and evaluate machine learning models focused on agricultural tasks such as weed and crop detection. Build and maintain datasets and data pipelines using image and sensor data collected under real field conditions. Enhance model robustness and efficiency to perform well across different crops, lighting, and weather scenarios. Integrate AI models into production software running on autonomous robotic systems. Collaborate with robotics, software, and product teams to address customer needs with technical solutions. Define experiments, establish evaluation metrics, and communicate results to guide development. Monitor and improve model performance through ongoing testing and debugging. Establish best practices for machine learning lifecycle including development, testing, versioning, and deployment.
Qualifications
Professional experience in developing and deploying machine learning models, with strong Python programming skills and familiarity with ML frameworks such as PyTorch or TensorFlow. Understanding of computer vision principles, experience working with image data, and performing model evaluations. Ability to manage projects from problem definition through experimentation and full production deployment. Commitment to writing maintainable, well-tested code and creating reliable real-world solutions. Strong English communication skills and a collaborative mindset. Curiosity and adaptability in a dynamic company environment. Desirable but not required are experience in robotics or autonomous systems, edge device deployment, MLOps, object detection, image segmentation, and familiarity with C++, Linux, ROS, or cloud infrastructure.