About me

I am a Ph.D. student advised by John Kitchin (previously by Zachary Ulissi) at Carnegie Mellon University. My work broadly focuses on building efficient machine learning methods and frameworks to accelerate molecular and material discovery. As part of my Ph.D., I have been collaboratively working on the Open Catalyst Project with Meta AI.

Research Contributions

Deep Learning - Diffusion models, Graph Neural Networks, Multi-task Learning, Transfer Learning

  • developing diffusion framework for catalysis. AdsorbDiff [ICML 2024]
  • developing an atomic foundation model that works across the domains of molecules, materials, and proteins. JMP [ICLR 2024]
  • developing novel sota graph neural networks for large-scale molecular and material datasets. SpinConv, SCN [NeurIPS 2022]
  • developing novel transfer learning method that helps with transfer across molecular domains. TAAG [JCP 2022]

Catalysis - DFT dataset, analysis

  • developing large-scale catalyst datasets focusing on energy applications. OC22 dataset [ACS Catalysis 2023].
  • analyzing errors, metrics, and approaches in training models for large-scale datasets. OC Challenges [ACS Catalysis 2022], Error Imbalance in OC20 [Catal. Sci. Technol. 2024]

Industrial Experience

I’ve been fortunate to do an internship across big tech, early-stage startup, and core industry.

  • Samsung (Summer’24)- I interned with Eric Wang at the Advanced Materials Lab of Samsung Semiconductors as an ML Research Scientist focusing on end-to-end discovery of solid-state battery electrolyte materials - from generative models to experimental validation.

    Key industrial takeaway: Problems in batteries and semi-conductor material synthesis

  • Orbital Materials (Summer’23) - I interned with an early-stage startup, Orbital Materials where I worked with the founding team (Mark Neumann, Jonathan Godwin) on developing a generative foundation model for material design focusing on the applications of porous materials. Additionally, learned about expts., startups, business, and more end-to-end about this field.

    Key industrial takeaway: ML engineering, merging ML with expts, learn about startups, business, and VCs.

  • Meta AI (Summer’22) - I interned with Meta AI and worked with Brandon Wood, Larry Zitnick. I primarily worked on conceptualizing and developing an atomic foundation model that works across the domains of molecules, materials, and proteins. Additionally, contributed to developing the Open Catalyst 2022 dataset, SCN model.

    Key industrial takeaway: Training AI models at large scale, ML engineering.

My undergraduate research experiences have been broadly related to computational aspects of Chemical Engineering, having completed projects in Deep Learning, Numerical Modeling, MD Simulation, and CFD applications. I have worked on numerous projects with Prof. Anurag Rathore at IIT Delhi in the field of modeling bioprocesses using CFD and Deep Learning. I have also been fortunate to do research internships with Prof. Faye McNeill, Columbia University and Prof. Duane Loh, the National University of Singapore developing machine learning methods for Air Quality and Protein Motif identification applications respectively.

Apart from research and academics, I was involved in various sports and cultural activities at IIT Delhi. Played water polo for the college team and was part of the Debating Club. In my free time, I read non-fiction novels (mind, human behavior, human body, economics), watch movies/TV shows/documentaries, run and swim, play chess (send me a challenge on chess.com), and play the keys.

If you have questions about my research or want to collaborate on anything, feel free to reach out to me via email.

Recent News