Seyyed Ali Ahmedi

At OpenGeoHub, Seyyed Ali works on machine learning, computer vision, and High-Performance Computing (HPC) for large-scale geospatial and Earth Observation applications. His work focuses on representation learning from long-term satellite data, large-scale model training and inference, and the deployment of open-weight language models on multi-GPU infrastructure.
His key responsibilities include:
Developing machine learning and computer vision methods for large-scale geospatial and remote-sensing datasets.
Designing and training representation-learning models for spatiotemporal Earth Observation data.
Building scalable HPC workflows for preprocessing, training, and inference over large satellite archives.
Optimizing distributed AI workloads across multi-GPU systems, including NVIDIA A100 infrastructure.
Deploying and evaluating open-weight large language models for research and agent-based workflows.

One of his current projects involves generating annual Landsat embeddings from OpenGeoHub’s approximately petabyte-scale archive of 16-day Analysis Ready Data extending back to 1997. This includes training a spatiotemporal embedding model from scratch and developing large-scale inference pipelines.
Previously, Seyyed Ali worked at the Sony R&D Center in Stuttgart, where he researched motion ghosting reduction for High Dynamic Range imaging and contributed to imaging technology related to Sony IMX sensors.