I am an Assistant Professor in the Artificial Intelligence Group at WUR and a member of ELLIS and 4TU+.AMI. My research focuses on two critical challenges in AI for science: Limited data and computational cost.
My expertise is in Scientic machine learning and physics-informed machine learning, with a focus on developing generalizable, data-efficient, and computationally efficient methods for sustainable engineering.
Prior to joining WUR as an Assistant Professor, I was awarded the Schmidt AI in Science Postdoctoral Fellowship at the University of Oxford. I obtained my PhD at TU Delft (2021–2024), focusing on physics-informed machine learning for complex systems. I then continued at TU Delft as a postdoctoral researcher (2024–2025).
I completed dual master’s degrees (fully-funded) in high-performance scientific computing at Université de Lille and in mathematics at South Asian University. I did my master internship at ETH Zürich (CAMLab). I obtained my bachelor’s degree in mathematics from the University of Delhi.
🇮🇳 Delhi → 🇫🇷 Lille → 🇨🇭 Zürich → 🇳🇱 Delft → 🇬🇧 Oxford → 🇳🇱 Wageningen
News
- Our workshop on Physics-Governed AI for Energy was accepted at AAAI 2027!
- Our paper on Towards ecologically meaningful Foundation models got accepted at Nature Machine Intelligence!
- I will give a keynote at DUCOMS on Towards Sustainable AI across Science in November 2026.
- I will be attending the Mathematics of Physics-Oriented Machine Learning workshop at the Mathematisches Forschungsinstitut Oberwolfach (MFO) in November 2026.
- I’ll be presenting on “Towards Sustainable AI for Engineering” at the European Commission’s Joint Research Centre (JRC) on 6 November.
- I am co-organising Parsimonious Scientific machine learning at the AAAI Symposium, November 2026.
- Our paper on understanding Plant protein using physics informed learning got accepted at Future foods!
- I was awarded the Early Career Award from Digital Futures, KTH for developing physics-informed machine learning methods for plant-based food applications.
- Our paper on Fast training of accurate physics-informed neural networks is selected for Oral presentation at ICLR 2026.
- Our paper on reducing computational cost through precision-induced learning in Physics-informed machine learning has been accepted for publication in Neurocomputing!