Astrophysics/Machine Learning PhD Researcher

I’m a PhD student at the University of Surrey.
I develop machine learning surrogate models to predict the dynamical evolution of astrophysical systems, with a focus on supermassive binary black holes in galactic mergers. My work combines graph neural networks, neural ODEs, and symbolic regression to build interpretable, scalable models for spatiotemporal systems.

Technical Skills

• Programming Languages: Python, SQL
• Machine Learning Frameworks: PyTorch, PyTorch Lightning, PyTorch Geometric, scikit-learn, Pyro, Hugging Face Transformers
• Scientific Computing & Data Handling: pandas, NumPy, SciPy, h5py, Jupyter
• Workflow & Experimentation: Git, Bash, Weights & Biases

Education

• PhD Astrophysics, University of Surrey (Sept 2023 - Current)
• MSc. Computer Science, Swansea University (2021 - 2022)
• MSci. Physics with Particle Physics and Cosmology, University of Birmingham (2016 - 2020)

Work Experience

• Freelance Data Analyst, TELUS (Aug 2022 - Aug 2022)
• Research Intern, Particle Physics group, University of Birmingham (Summer 18’ and 19’)

Current Projects

Machine Learning for Predicting the Time Evolution of Supermassive Black Hole Binaries_ (Ongoing)

Workshops

• International workshop on diffusions in machine learning: foundations, generative models, and optimisation, participant

Research Interests

I develop surrogate models for predicting the dynamical evolution of binary black holes in N-body simulations of galaxy mergers, using advanced machine learning techniques such as Neural ODEs, Graph Representation Learning, and Symbolic Regression. My current work emphasises the use of pretraining and transfer learning on related spatiotemporal systems to improve both accuracy and computational efficiency.