Description

You will develop AI-based perception and spatial understanding for inspection drones, combining 2D images and 3D point clouds into robust semantic representations of complex indoor environments.

Tasks

• Design and implement AI models and algorithms for 2D/3D scene understanding
• Combine camera and point-cloud data into consistent semantic spatial maps
• Develop semantic understanding to support mapping, navigation and inspection planning
• Prepare and analyze real-world drone datasets
• Evaluate models using quantitative metrics, visualizations and targeted experiments
• Build Python/PyTorch research and production pipelines from experimentation to deployment
• Collaborate with mapping, sensors and product teams on spatial AI capabilities

Profile

• 3+ years of experience developing machine learning or computer vision systems
• Deep expertise in semantic segmentation in 2D imagery and/or 3D point clouds
• Proven experience developing production-grade computer vision systems
• Strong proficiency in Python and PyTorch
• Strong understanding of the end-to-end ML lifecycle
• Experience with real-world datasets, evaluation and systematic error analysis
• Proficiency in English; French is a plus

Offer

• 100% workload
• Starting date as soon as possible
• Flexible schedules with option to work remotely up to 2 days per week
• 3-6 months interim, then a permanent position

Company

Robotics / drone technology