Marijn Valk is a generalist Software Engineer specializing in Data & ML with nearly a decade of experience building production systems in financial and healthcare domains. He has led design and implementation of petabyte-scale, multimodal data platforms and ML-driven clinical assistants, and helped shape engineering practices as the first hire at kaiko.ai. Marijn combines backend and MLOps expertise—contributing to prominent open-source projects like Dagster and MLflow—to improve orchestration, storage integrations, and UX of ML tooling. Comfortable across the stack, he has a track record of operationalizing advanced analytics at banks and hospitals, and he brings a pragmatic security-aware mindset from roles like Security Champion. Based in Zurich, he pairs research-informed thinking from his Utrecht Masters with hands-on shipping of scalable, auditable systems.
9 years of coding experience
4 years of employment as a software developer
Master's degree Business Informatics, Master's degree Business Informatics at Utrecht University
Open source platform for the machine learning lifecycle
Role in this project:
MLOps Engineer & Back-end Developer
Contributions:22 reviews, 13 commits, 14 PRs in 11 months
Contributions summary:Marijn primarily focused on enhancing the MLflow tracking UI and backend functionalities. They implemented features like a duration column in the runs table and a lifecycle stage display, improving the user experience. Furthermore, they added and improved the CSV export capabilities by including start time and duration metrics. Additionally, they worked on integrating Azure Blob Storage with the MLflow tracking server, and contributed to several bug fixes.
An orchestration platform for the development, production, and observation of data assets.
Role in this project:
Back-end Developer
Contributions:4 reviews, 31 PRs, 15 comments in 1 year 3 months
Contributions summary:Marijn primarily focused on enhancing the `dagster-deltalake` library by addressing a configuration issue related to storage backend options. They implemented changes to ensure that the library correctly handles configuration values of various types, preventing type errors within the `DeltaTableResource` component. Their work involved modifying config classes and updating tests, specifically within the context of Azure storage, to ensure the correct behavior. Furthermore, they added functionality to the `dagster-github` integration by adding methods for branch creation and pull request creation.
operationpythonobservationschedulermetadata
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.