About
I'm a NASA Postdoctoral Program (NPP) Fellow at the Jet Propulsion Laboratory. Here, I work on several initatives: building a transformer-based emulator for ocean world characterization at Europa, the evolution of the solar wind and interplanetary magnetic field beyond 1 AU, and plasma and atmospheric conditons at Saturn's moon Enceladus.
I'm a science affiliate on the Europa Clipper mission, through the Europa Clipper Magnetometer (ECM) instrument, and a co-founder of JPL's Model Understanding Surrogates and Emulators (MUSE) task force. I am also involved in mission formulation for NASA's explorer class, and review for NeurIPS and JGR.
My work sits at the intersection of AI and space exploration — spanning science, engineering, and the systems that will define the next generation of missions.
Background
I did my undergraduate in Computing with Business at Oxford Brookes Unversity, then spent a few years in industry; working at the NHS followed by the circuit board industry as a process and design engineer.
I came back to academia for an MSc in systems engineering with space systems at the UCL Mullard Space Science Laboratory (valedictorian, 2019), then a PhD in space physics and applied machine learning, also at UCL (2024). During my PhD I was a visiting scholar at Caltech, a research intern at NASA's Jet Propulsion Laboratory, and I received the 2023 UCL Alan Johnstone Award for Outstanding Scientific Achievement. I also won 2x team achievement awards for my work on the CIRCE and Phoenix CubeSat missions, the former of which was part of the first space mission to launch on UK soil.
Between the PhD and JPL I held a short postdoctoral position at the National Institute of Polar Research (国立極地研究所) in Japan, working on ML for auroral physics, magnetosphere-ionosphere coupling, and citizen-science observations.
Finally, I was a mentor on the Orbyts programme, which partners researchers with schools to develop scientific research skills. The programme specifically aims to widen participation in STEM for students from historically under-represented backgrounds, particularly in low-income areas and among young women.
Research Interests
- Machine learning for scientific discovery
- AI co-scientists and agentic research workflows
- Neural surrogates and emulators for physics simulation
- Uncertainty quantification and validation in scientific ML
- Space plasmas, ocean worlds, magnetospheres, and the search for life in our universe.
Why AI/ML?
AI is arguably the most consequential general-purpose technology since the harnessing of electricity or fire. Most breakthroughs are vertical—useful within a single domain—but AI is horizontal and compounds across every field that adopts it, including space science and engineering. As such, our community cannot simply consume AI; we must sit at the forefront of it, building our own ML models rather than waiting for them. That said, we must not use AI to replace physical understanding, but rather to reach and validate it faster, at time scales neither a human nor a classical solver can match alone.