At Kaiko we train State-of-the-Art medical foundation models that will power the next generation of medical AI products, thus realizing the AI healthcare revolution. We bring together large multimodal medical datasets in a scalable cloud-based way. This allows AI algorithms to learn and uncover novel relations that can advance medical research, and thereby lead to better treatment options and improved patient outcomes.

As Principal Machine Learning Scientist, I contributed to training the Transpara product evaluated in the large MASAI breast cancer randomized control trial in Denmark reported by Kristina Lång in The Lancet. This study found that our AI product reduces radiologist workload by 44%, while detecting 20% more cancer cases. This forms a milestone in medical AI, recognized by Nature Medicine as one of the Notable advances of 2023. My focus was on innovating the Deep Learning technology driving ScreenPoint's Transpara product for breast cancer detection with the ML team. Over a period of three and a half years, we brought this technology to the point where it's now viewed worldwide as best in class, surpassing human radiologists. This contributed to the use of ScreenPoint's Transpara in daily medical practice in Denmark's national breast cancer screening programme, and a 28M$ investment from New York VC fund Insight Partners.

As a mentor in several editions of RockStart's AI Accelerator programme, I evaluated many AI startups and coached and advised a selection of AI startups, including Birds.AI, Medvice, Skinive, and Plantik.

IMC is a leading global market maker, and became Designated Market Maker (DMM) on the New York Stock Exchange (NYSE) in 2014.As Principal Machine Learning Researcher, I contributed to introducing machine learning into IMC's trading practice.

Leading the development of predictive analytics technology and products within Quintiq.

Quintiq's supply chain planning and optimization software solves many of the world's most complex planning puzzles. Led the development of Quintiq's Demand Planner into the mature solution that makes up part of Quintiq´s innovative S&OP offering today.

Founded and developed Adapticon, a machine learning startup. Adapticon is one of the earliest companies that developed and deployed Deep Learning algorithms. Forecasting and predictive analytics projects for major Dutch energy companies (Essent, Nuon), Booking.com, Aldipress; traffic prediction projects for all three global leaders in digital maps.

Research into machine learning, coevolution, bioinformatics.

Postdoctoral research on coevolutionary algorithms at Jordan Pollack's DEMO Lab; Fulbright scholarship for Research and Lecturing..

PhD research at the VUB Artificial Intelligence Lab