I'm Amir Masoud (Masoud) Nourollah, a PhD student at Cardiff University, working across the Cardiff NLP group and the Knowledge Representation and Reasoning (KRR) group in the School of Computer Science and Informatics.
- 🔬 My PhD research focuses on explainability and interpretability of deep learning models for molecular dynamics and machine-learning interatomic potentials (MLIPs).
- 🧠 My MSc research applied deep learning to neural image segmentation — detecting neurons and quantifying their morphology in fluorescence microscopy images.
- 💼 Before academia, I worked as a software and machine learning engineer, building systems across computer vision, speech, and NLP.
- 🌍 Outside research, I enjoy traveling, cooking, and exploring new corners of Linux and open-source tooling.
|
Equivariant networks, graph neural networks, and symmetry-aware architectures for structured scientific data. |
Explainable, generalisable surrogate models for atomistic simulation and molecular dynamics. |
Applying ML-driven materials discovery to accelerate next-generation energy storage research. |
| Project | Description | Status |
|---|---|---|
| 🧪 GOAL | General Open Atomistic Laboratory — a Python framework for training machine-learning interatomic potentials (MLIPs) on atomistic systems | 🟢 Active |
Older repos (SCAI, GMD-26, segmentation projects, etc.) are archived / exploratory and no longer under active development — GOAL is where the current work lives.
"Making deep learning models more transparent and accessible for scientific discovery."
📫 Reach me via nourollah.me · LinkedIn · Google Scholar