PhD researcher in robotics at Khalifa University. I build AI systems that help robots and sensors make sense of the real world, from the seabed to satellite images.
I work at the meeting point of robotics, computer vision and machine learning, with a focus on systems that have to work outside the lab: with little data, limited hardware and real consequences when they get things wrong.
I am a member of the robotics group of Prof. Jorge Dias at Khalifa University, co-supervised by Dr. Federico Renda. Before my PhD, I studied mechatronics and artificial intelligence at the University of Sheffield.
Computer vision that runs on robots and embedded hardware, and knows when to say "I'm not sure".
Models that generalise from a handful of examples, and honest ways to test whether they really do.
Robots that decide what is worth reporting, when bandwidth, power and time are all limited.
An AI pipeline that finds the Endangered coach whipray in underwater photos, rejects look-alike animals, and names the species only when confident. Built from a few hundred public images, without training a large model.
My PhD project: instead of streaming video over slow acoustic links, the robot recognises what it sees and sends short, structured reports, with a metric for how faithfully those reports match reality.
A study of why species-recognition models fail on new footage, showing that the number of distinct recordings matters more than the number of training images.
Team lead in the UAE Space Agency Space Academy hackathon, working on early detection and impact assessment of floods, droughts, dust storms and wildfires from satellite time series.
A lab project combining underwater simulation and a Meta Quest 3 headset to explore and assess coral reef health in mixed reality.
I am always happy to talk about robotics, AI and research collaborations.