male, 27
I was born in Jabalpur, India. I have always been fascinated with computers and enjoy applying machine learning to real world problems. This website is a collection of my current research interests, past collaborations, ramblings and when and where you might be able to find me IRL.
I recently completed a master's degree in Machine Learning at the University of Tübingen and am currently actively applying for PhD positions. I am broadly interested in using AI to enable scientific discovery, with a particular affinity for applications in healthcare.
Scientific discovery and invention is what makes us humans. I have always used AI as a means to an end and my current research focus is to use it for scientific discovery. Researchers already use AI for their day to day work, but can we build tools that accelerate this even further? How would you evaluate such a tool? Do we have benchmarks for evaluating capabilities needed for scientific discovery?
I would love to hear from you if you have some ideas that you want to explore. Just shoot me an email explaining the idea in 2 paras. :)
Most recently, I completed my master's thesis in Robust Machine Learning Group at ELLIS Institute Tübingen focusing on designing and evaluating AI-powered tools for teachers in education.
Before my master's I focused on deploying and applying computer vision techniques in a not-for-profit, Wadhwani AI. I later joined Rephrase AI (now acquired by Adobe), a Series A startup developing virtual avatars, where I worked with generative models, including variational autoencoders, for synthetic speech generation.
During my bachelor's, I was part of Sustainability Lab, IIT Gandhinagar where I used active learning for intelligent air quality sensor placement for optimal coverage at a budget.
I’m happy to help people find relevant work, groups, or researchers in areas I know. If you think an introduction might be useful, feel free to reach out with some context.
A selection of publications and projects spanning deployment-aware AI, scientific computing, and tools built for constrained real-world settings.
A framework for evaluating deployed AI systems in low-resource settings, where noisy inputs, code-switching, intermittent connectivity, constrained hardware, and human oversight matter alongside model quality.
An open pest-image dataset gathered over five years through a mobile advisory system used in Indian cotton fields, preserving the scale, noise, and variation of an actual agricultural deployment.
An interactive introduction to Gaussian-process surrogates and acquisition functions, showing how sequential decisions can optimize expensive black-box systems more efficiently than exhaustive search.
Literature reviews, research notes, and web-native explanations of ideas I am studying, alongside informal writing preserved from my earlier website.
Six years across applied AI, startups, Germany, and a master's degree taught me to put curiosity back in the driver's seat.
A 2020 note about joining Wadhwani AI as a Research Fellow and working on AI for social good.
Reflections on discomfort, gratitude, and a snow trek in Uttarakhand in March 2020.
Recent updates and where you can find me next.
conference
I will be attending Thinking About Thinking's annual meeting on open problems for AI.
education
My master's defence is scheduled for 11 am.
community
A year-long programme bringing early-career researchers together around interdisciplinary work on intelligence.
education
Evaluating AI-Supported Planning for Differentiated Mathematics Homework, completed with the Robust Machine Learning Group.
career
Concluded my role as an AI Research working student.
career
Joined as an AI Research working student.
education
Joined the University of Tübingen as a master’s student.
talk
Gave a guest lecture for the Probabilistic Machine Learning course.
publication
A workshop paper about machine-learning deployment challenges at Wadhwani AI.
career
Joined as a Deep Learning Researcher.
talk
conference
career
Joined as a Research Fellow.
publication
career
No current or upcoming entries.
No recent entries.
No archived entries.
