AI Research Associate: Physics-Informed Env Forecasting

University of Leicester; Leicester; On-site; GBP 37,000 - 45,000; Full time

Job summary

University of Leicester invites applications for a Research Associate in AI for Environmental Forecasting (SENTINEL project) to lead the AI development strand of a UKRI-funded effort building a physics-informed AI forecasting system for chemical fugitive emissions.

You will design, train and evaluate the Receptor Transfer Function Learner, assemble multi-modal data, and work with GPU/HPC facilities while collaborating with UK Health Security Agency and Met Office.

Qualifications

• PhD or equivalent in a quantitative discipline with track record in ML/DL.
• Experience developing ML/DL models with real-world time-series data.
• Strong Python programming and experience with PyTorch, TensorFlow or JAX.
• Experience contributing to peer-reviewed publications or equivalent research outputs.
• Ability to work independently and in a collaborative research team; clear communication.
• Experience with graph neural networks, diffusion models, physics-informed ML, weather/geospatial data, HPC advantageous.

Responsibilities

• Lead the AI development strand of SENTINEL project.
• Build a multi-modal training dataset and design, train and evaluate the Receptor Transfer Function Learner.
• Benchmark SENTINEL against established dispersion models and publish results.
• Collaborate with partners including UK Health Security Agency and Met Office.
• Contribute to high-quality research publications and dissemination.
• Utilise GPU clusters or HPC resources for experiments.

Skills

Graph neural networks

Generative models

Diffusion models

Physics-informed ML

Explainable AI

Weather data

Geospatial data

GPU / HPC experience

Education

PhD or equivalent in a quantitative discipline

Tools

Python

PyTorch

TensorFlow

JAX

NumPy

pandas

xarray

Job description

University of Leicester invites applications for a Research Associate in AI for Environmental Forecasting (SENTINEL project) to lead the AI development strand of a UKRI-funded effort building a physics-informed AI forecasting system for chemical fugitive emissions.

You will design, train and evaluate the Receptor Transfer Function Learner, assemble multi-modal data, and work with GPU/HPC facilities while collaborating with UK Health Security Agency and Met Office.