Keith Butler

Keith is Associate Professor in Computational Materials Chemistry. Previously he worked as a staff scientist at the Rutherford Appleton Laboratory, as a post-doctoral resaercher in the groups of Aron Walsh and John Harding, flipping cheestakes in 99 Miles to Philly and the man behind the burger vending machine in Dublin’s famous Ilac Centre.

Keith serves as an associate editor of npj Computational Materials and on the editorial board of Machine Learning Science and Technology. Keith is an active developer of several open source materials design packages SMACT, SuperResTomo, MacroDensity and a strong advocate of open science.

Bradley Martin

Bradley is a Research Assistant in the MDI group at UCL.

His research aims to develop physics-informed deep learning models to predict polarisation and shift currents of semiconductors and to aid in discovery of new ferroelectric and photovoltaic materials with good shift currents and high polarisation.

Previously, during his masters in theoretical physics at Imperial College London, Bradley studied BRST quantisation and bosonic string theory in Prof. Dan Waldram’s group. Bradley then transitioned towards research with more… tangible outputs, and did his PhD with Dr Jarvist Frost and Prof. Jenny Nelson at Imperial College London, developing new variational path integral methods for modelling charge-carrier mobility and optical absorption of polarons in polar semiconductors. Here he also co-developed the Julia package PolaronMobility.jl.

When not lost in his own mind, Bradley enjoys playing Dungeons & Dragons and boardgames, hiking up mountains (especially in Wales), losing at pub quizzes and trying (with questionable success) to learn French.

Matthew Walker

Matthew is a PhD student in the group at UCL. His project aims to use graph neural networks to predict optical and electronic properties of semiconductors from chemical composition. During his master’s in materials modelling (also at UCL) he studied sustainable thermoelectric materials in David Scanlon’s group, before a brief stint in the corporate world confirmed his suspicions of its cold cynicism and his desire to return to the academic world. When he isn’t stuck behind a computer, he likes cooking, attempting a spectrum of exercise modalities, and learning new languages (currently Italian and Kotlin). Matthew’s webpage

Mueen Taj

Mueen is a PhD student in the group at UCL. His project is based on developing machine learning approaches to enhance the analysis of neutron spectroscopy experiments, bridging the accuracy of density functional theory with the scalability of molecular dynamics for studying complex materials. Prior to his PhD, he studied materials chemistry at the University of Bradford and briefly ventured into the banking sector, before realising his passion for academic research leading to his big move to London. Outside of his work, Mueen is an avid gamer with a growing interest in metroidvania games, with Hollow Knight being a personal favourite.

Cyprien Bone

Cyprien is developing a large language model to explore a wide range of crystal structures for renewable energy applications. The goal is conditional structure generation, allowing researchers to design materials with specific properties or constraints in mind. This project builds on the work of Luis Antunes’ CrystaLLM. After a year in industry at an agrochemical research park outside of London—where he realised lab work wasn’t his calling—his master’s thesis focused on studying Graph Neural Networks to predict opto-electronic properties in organic photovoltaics. This sparked his love for computers and renewable energy research, finding his way to help carbon reduction initiatives. Outside of work, Cyprien enjoys baking bread, exploring global cuisines, rock climbing and hiking up mountains (or hills as they call them in England).

Weihang Xie (David)

Weihang (David) is a visiting student from the National University of Singapore.

Junayd Ul Islam

Junayd is an MSci student at UCL. His growing interest for materials science reached a zenith when studying the chemistry of photovoltaics. Junayd capitalised on the relationship he formed with the lecturer of that module, Professor Butler, to perform a research project in the topic. Junayd hopes to refine the similarity learning method used in material science by evaluating the possible vector spaces and distance matrices; this will inform a powerful machine learning method which can identify new crystal structures for use for PV technology. In his spare time, Junayd loves to read, write and perform poetry.

Jumanah Lazumi

Jumanah is a MSci at UCL. Her project looks at rare, yet impactful Black Swan events and the likelihood of discovering a greatest of all time (GOAT) material with properties far better those found in the average material. The aim behind Jumanah’s project is to examine a few materials properties and determine how likely it is to discover a GOAT material. These findings could motivate new research and inform where resources should be directed. In her spare time, Jumanah enjoys reading science fiction and baking all types of desserts.

Previous Members

• Masaki Hiratsuka (visitor from April 2023 - March 2024) - Assoicate Prof. Kogakuin University
• Irina Stanojevic (visitor in June 2024) - PhD candidate - University of Belgrade
• Jaivin Gohil (summer student June - Aug 2024)