At OpenGeoHub, Jay supports the design, optimization, and maintenance of High-Performance Computing (HPC) and data infrastructures. He is in charge of developing scalable solutions that bridge the gap between complex machine learning applications and the physical hardware required to power them.
His key responsibilities include and are not limited to:
- Architecting HPC environments, including the deployment of distributed computing systems like Slurm and Hadoop.
- Managing scalable back-end solutions and file services such as S3, Ceph, and SeaweedFS to handle massive datasets.
- Optimizing data storage and performance through advanced compression methods and binary formats like Parquet.
- Monitoring hardware lifecycle and energy efficiency, ensuring continuous and reliable operation of server clusters.
- Processing geospatial time-series data, facilitating the use of Earth Observation imagery in large-scale modeling.
- Fostering technical collaborations with international research networks and project partners.
Jay previously served as a Researcher at the Center of Expertise in Big Geodata Science (CRIB) within the Faculty of ITC at the University of Twente. With this strong foundation in big geodata and remote sensing infrastructure, he now serves as a Computer Scientist at OpenGeoHub. Besides his core technical duties, his function involves translating research requirements into high-performance computing realities to support global scientific outputs.