Adeesh Kolluru

prof_pic.jpg

I am currently leading AI at Radical AI, where we are working on combining AI, computational chemistry, and autonomous labs to accelerate materials discovery. Previously, I was a Ph.D. student advised by John Kitchin and Zachary Ulissi at Carnegie Mellon University. My work broadly focuses on building efficient machine learning methods and frameworks to discover novel materials. My research interests span multiple areas including diffusion models, graph neural networks, multi-task learning, and more recently LLMs. You can find more detailed information in publications and CV.

During my PhD, I primarily contributed to the Open Catalyst Project in collaboration with Meta AI, developing large-scale models and datasets for computational chemistry. I also gained valuable industry experience through internships at Samsung Semiconductors working on battery materials discovery, Orbital Materials developing generative models for porous materials, and Meta AI advancing foundation models for atomic systems. These experiences have shaped my understanding of different perspectives on AI-driven materials discovery.

Recent News

πŸš€ Apr 2025: Released EGIP - state-of-the-art interatomic potential achieving best performance on thermal conductivity prediction!

⚑ Apr 2025: Launched TorchSim - next-generation PyTorch-native atomistic simulation engine with 100x speedup over ASE!

πŸš€ Oct 2024: Joined Radical AI

πŸŽ“ Oct 2024: Successfully defended PhD thesis at Carnegie Mellon University!

πŸŽ‰ May 2024: AdsorbDiff is accepted at ICML 2024!

πŸ“„ Oct 2023: Our Joint Multi-domain Pretraining JMP across 3D atomic datasets is out.

πŸš€ May 2023: Interning at Orbital Materials.

🌍 Aug 2022: Mentoring at Tackling Climate Change with ML workshop, NeurIPS'22.

πŸ“Š Jun 2022: Spherical Channel Network (SCN) paper is up.

πŸ“ˆ Jun 2022: OC22 Dataset for oxide electrocatalysis is up.

πŸ† Jun 2022: Open Catalyst Challenge is announced for NeurIPS 2022!

πŸ“ Jun 2022: Perspective on challenges with training ML models for materials is up.

🎀 May 2022: Gave a talk on Transfer Learning at Toyota Research Institute (TRI) workshop!

πŸ“„ Apr 2022: TAAG paper is published.

πŸ€– Mar 2022: Will be interning at Meta AI this summer!

πŸ… Feb 2022: Received Phillips and Huang Family Fellowship in Energy [press]

πŸ“š Jan 2022: OCP Tutorial as part of Climate Change AI Workshop at NeurIPS

🎀 Nov 2021: Gave a talk at AIChE 2021 on Transfer Learning with GNNs on Molecular Datasets

πŸ“„ Jun 2021: SpinConv paper is up on arxiv

πŸ† Jun 2021: Open Catalyst Challenge announced for NeurIPS 2021

πŸŽ“ Jun 2021: Excited to be selected for LOGML Summer School!

Personal Interests

Beyond my research activities, I maintain diverse interests that contribute to my well-rounded perspective. During my undergraduate years at IIT Delhi, I was actively involved in sports and cultural activities, including playing water polo for the college team and participating in the Debating Club. These experiences have enhanced my teamwork and communication skills, which prove valuable in collaborative research environments.

In my personal time, I enjoy reading non-fiction literature on topics ranging from cognitive science and human behavior to economics and technology. I also stay engaged with current developments through documentaries and maintain an active lifestyle through running and swimming. As a chess enthusiast, I regularly play online (feel free to challenge me on chess.com) and enjoy playing piano as a creative outlet.

Let’s Connect!

πŸ’¬ Get in Touch

If you have questions about my research, are interested in collaboration opportunities, or would like to discuss potential projects, please feel free to reach out via email.