Research Fellow (Urban Microclimate Modeling and Graph Neural Networks)
NATIONAL UNIVERSITY OF SINGAPORE
Singapore, Singapore
Not specified
Graph neural network (gnn) development
Spatiotemporal deep learning implementation
Python programming with pytorch or tensorflow
The National University of Singapore is seeking a Research Fellow specializing in Urban Microclimate Modeling and Graph Neural Networks. The role involves developing advanced modeling frameworks and conducting extensive data collection and validation to enhance urban climate understanding, with a strong emphasis on collaboration and research dissemination
Job Summary
The role involves designing a hybrid physics-AI framework to generate high-resolution ambient air temperature maps from satellite-derived Land Surface Temperature data.
Candidates will execute field validation campaigns across diverse HDB precincts using sensor deployment, drone-based measurements, and mobile sensing data.
The successful applicant will package the validated model as a deployable module compatible with HDB's Integrated Environmental Modeller and engage with government agency partners.
Matching Summary
Match Score: 85
The National University of Singapore is seeking a Research Fellow specializing in Urban Microclimate Modeling and Graph Neural Networks. The role involves developing advanced modeling frameworks and conducting extensive data collection and validation to enhance urban climate understanding, with a strong emphasis on collaboration and research dissemination.
Skills & Requirements
Must-have
Graph Neural Network (GNN) development
Spatiotemporal deep learning implementation
Python programming with PyTorch or TensorFlow
Remote sensing data processing
Geospatial analysis tools proficiency
Peer-reviewed publication track record
Nice-to-have
Experience with ENVI-met simulation tools
Knowledge of life-cycle carbon assessment
Familiarity with UHI effects and thermal comfort
QGIS/ArcGIS plugin development skills
Collaboration with multi-institutional teams
Key Requirements
PhD in Computer Science, Urban Building Science, or related quantitative field
Demonstrated experience with Graph Neural Networks and spatiotemporal deep learning
Strong Python programming skills with deep learning frameworks
Proven track record of peer-reviewed publications in machine learning or remote sensing