Research & Education

Maarten Stol is the Chief Scientific Officer at BrainCreators, where he excels at bridging the gap between cutting-edge AI research and the industry's demand for scalable software products.

He earned his M.Sc. in Artificial Intelligence from the University of Amsterdam in 2000, with a focus on computational logic and nature-inspired optimization algorithms. His early interests spanned from automated theorem proving to genetic algorithms and neural network technology.

Joining BrainCreators in 2016, Maarten has since delved into a wide range of Machine Learning research topics at the intersection of industry and academia. His work on areas like object detection models, deep clustering, and neural network compression has directly led to new product features, exciting academic thesis projects, and publications, successfully attracting young talent to the company. Today, he works closely with BrainCreators' ML teams, offering guidance on experiment design, ML deployment monitoring, and developing their technical skills.

Before BrainCreators, Maarten gained diverse experience as a teacher, researcher, consultant, and software developer, often within innovative start-ups. Now, he advises industry partners on AI adoption, explainable AI, and aligning AI potential with client needs. He regularly shares his insights on the interplay between AI academia and industry through speaking engagements and contributions to BrainCreators' "Insights" blog.

Committed to knowledge sharing, Maarten became a lecturer for the Master of Applied AI at the Amsterdam University of Applied Sciences (HvA) in 2023. Further deepening his academic pursuits, he joined the "Learning and Reasoning" group at the Vrije Universiteit Amsterdam (VU) in 2024 as a guest researcher. There, he is pursuing his Ph.D. in Neuro-Symbolic AI, happily reuniting his foundational love for logic and knowledge engineering with his professional experience in deep learning. His current academic research focuses on integrating symbolic knowledge into differential Machine Learning optimization and addressing challenges in Neuro-Symbolic benchmarking design.