I'm Roland Meertens, a machine learning engineer based in London. I love robotics — especially self-driving cars — and all the challenges that come with teaching machines to understand the world around them. In my spare time I build robots, weird gadgets, and side projects, and write about them on this blog.
You can also find my full CV here (PDF).
Leading geo and platform expansion for Wayve's autonomous driving stack. Responsible for scaling end-to-end deep learning systems to new cities and driving domains, coordinating across perception, planning, and infrastructure teams.
Built and deployed machine learning models for trust & safety, including top-of-the-world face verification and recognition systems. Also worked on features to add more AI capabilities to improve the dating experience. Wrote about efficient dataset creation on the Bumble Tech blog.
Designed and built products that help perception engineers evaluate and improve their autonomous driving models. Defined product roadmaps for lidar and camera annotation tools used by major OEMs. Bridged the gap between ML research teams and product delivery.
Developed deep learning models for 3D object detection and prediction using lidar and camera fusion. Built systems to determine what surrounds the vehicle and predict traffic participant behaviour. Rode along in a self-developed self-driving car through Munich city traffic.
Developed neural machine translation and automatic post-editing systems using sequence-to-sequence models. Presented results on deep learning for NLP at the GALA conference for language professionals.
Developed computer vision algorithms for autonomous micro air vehicles. Started a team to do autonomous drone racing — although I'm not sure you can call it racing if you fly 3 kph. Also worked on a flapping-wing ornithopter that navigates corridors using onboard vision, and a drone that autonomously searches for packages.
Developed social interaction behaviours for the Romeo humanoid robot, a 1.4m tall assistant robot built by Aldebaran Robotics (now SoftBank Robotics). See a 12-year-younger version of me in action.
Specialisation in cognitive artificial intelligence and brain-computer interfacing. Thesis on how to create a social dialogue system for social robotics — yes, this was before LLMs were a thing.
Started studying AI before anyone knew what AI was. Graduated bene meritum.