I'm Aiyush Gupta, a Materials Science student at Oxford working on semiconductor devices, robotics, and the software that connects them.

I have worked on semiconductor devices, with a particular interest in what we can learn by modelling & fabricating them.

At Oxford's Electronic and Interface Materials Laboratory, I've been building electrothermal models in COMSOL for TacOSPV, studying how heat and electrical behaviour interact in solar cells. This means working between detailed device physics and models simple enough to be useful at larger scales.

Alongside this, I worked on physics-constrained parameter extraction for silicon solar cells. A good fit can hide a lot of uncertainty: different parameter combinations can describe the same measurement. I want to know which properties we can reliably extract, which remain ambiguous, and what we would need to measure next.

I co-founded Oxford Robotics Club, bringing together over 100 members in our first term, and helped organise the Oxford Physical AI Hackathon.

I've also contributed to government roundtables on how the UK can leapfrog in robotics, and spoken on an Engineering Professors' Council panel for the Complex Systems Toolkit.

My own projects have taken me from a robotic gardener nominated for the UK TeenTech Awards to an experiment I ran aboard the International Space Station, using Earth imagery to build machine-learning systems that predict vegetation change.

With Pathway, I led a team that turned an NHS walking frame into a low-cost mobility-aid prototype in 36 hours, combining spatial sensing, hazard detection and an accessible interface. We won the Oxford Venture Hackathon.

I want to bring that same integration of sensing, software and hardware into experimental science. I'm interested in the machinery behind an autonomous lab: instruments that can be controlled in code, measurements traceable to how a sample was made, and experiments that a system can reproduce and build on.

Thin-film fabrication is where I'd like to start, connecting deposition conditions to measured properties in a loop that improves with each run.