Summary
Fabiola Maffra is a computer vision research engineer with over a decade of hands-on experience bridging academia and industry, now focused on vision-based navigation for autonomous UAVs in GNSS-denied environments. With a PhD from ETH Zürich and a track record of securing multi-year Amazon research awards, she combines deep learning for semantic segmentation, object detection, depth completion and feature matching with practical SLAM and navigation systems. Her work has driven real-world improvements in localization under challenging viewpoints, lighting changes and perceptual aliasing, and she has engineered lightweight solutions for resource-constrained drones. Earlier roles span real-time 3D graphics for the oil & gas sector, large-scale simulator refactoring that cut runtimes from a year to minutes, and AR camera-tracking libraries—showing a rare blend of systems optimization and advanced perception. Based in Zurich, she routinely translates synthetic data generation and style transfer techniques into robust visual localization gains. Colleagues value her ability to move projects from novel research concepts through full software lifecycle delivery in collaborative, cross-disciplinary teams.
10 years of coding experience
10 years of employment as a software developer
Pontifical Catholic University of Rio de Janeiro
Doctor of Philosophy - PhD, Computer Vision for Robotics, Doctor of Philosophy - PhD, Computer Vision for Robotics at ETH Zürich