I work at the Allen Institute for Cell Science, where I build computational methods to help scientists understand how cells organize, behave, and change over time. My work sits at the intersection of computer vision, deep learning, and cell biology, taking 3D microscopy images of cells and turning them into structured, queryable knowledge.
Background
I came to this through a winding path. I studied Computer Engineering at Pune Institute of Computer Technology, then worked as a Research Engineer at CNRS CINAM in Marseille, building automated pipelines for detecting and classifying red blood cells from blood flow recordings, a prototype for the ICOVELL spin-off.
I spent time as a visiting scholar at EMBL Heidelberg in Anna Kreshuk’s group, working on automated localization of kinetochores, proteins critical to cell division. This gave me my first real taste of computational biology at the cutting edge.
For my M.Sc. at Simon Fraser University, I worked in the Medical Image Analysis group under Prof. Ghassan Hamarneh, studying the structural analysis of the endoplasmic reticulum as part of the NanoscopyAI collaboration with Prof. Robert Nabi at UBC. My thesis developed network-based methods for analyzing ER dynamics, an approach I’m still building on.
Along the way, I’ve collaborated across domains: astrophysics with DeepSkies and DarkMachines, climate science with ECMWF through the ESoWC program, and particle physics through Google Summer of Code at CERN-HSF.
What I Care About
I believe the most impactful AI work in the next decade will happen at the boundary between machine learning and the sciences. Not because the ML is harder, but because the problems are richer, the data is messier, and knowing whether a model’s output is scientifically meaningful requires a kind of judgment that can’t be automated.
I’m drawn to Focused Research Organizations, charter cities, and the longevity space. I spend too much time on various Substacks and maintain a rabbit hole of reading material that I’ll never finish.