Research Associate - Machine Learning for Earth Observation

About the role

The School of Civil and Environmental Engineering (CEE) at NTU is looking for a Research Associate in Machine Learning for Earth Observation. The role develops self-supervised, temporal and multimodal models for satellite image time series and geospatial data, applied to coastline change, land subsidence, urban expansion and vegetation monitoring. Candidates can also explore emerging directions such as vision-language models, agentic AI and generative models. High-performance GPU computing and close supervision are provided.

Key responsibilities

• Develop deep learning models for satellite image time series and multimodal Earth observation data.
• Develop next-generation geospatial foundation models, such as carbon minimization.
• Explore emerging methods including vision-language and multimodal large language models linking satellite data with text, agentic AI workflows for geospatial analysis, and generative (e.g. diffusion) models for data synthesis.
• Apply the models to coastline change, land subsidence, urban expansion and vegetation monitoring.
• Build scalable, reproducible processing pipelines on high-performance and cloud computing, and maintain open-source code and datasets.
• Prepare data and reference labels, evaluate models, write papers and reports, present results and support the research group.

Required profile

• Master's degree in computer science, electrical engineering, remote sensing, geomatics, civil and environmental engineering or a related field.
• Knowledge of machine learning and deep learning.
• Experience with remote sensing or geospatial data; knowledge of SAR interferometry is a plus.
• Good communication skills, able to work independently and in a team.
• Self-driven and interested in pursuing a PhD.

Required skills

• Proficiency in Python and PyTorch on Linux/GPU computing.
• Experience with foundation models, vision-language models or AI agents is a plus.