Animashree (Anima) Anandkumar is the Bren Professor of Computing at the California Institute of Technology (Caltech) and a senior director of machine learning research at NVIDIA. She is also a co-founder of Accelerated Understanding, a company focused on AI education and research. Her research spans tensor-algebraic methods, deep learning, and non-convex optimization, with applications in scientific domains such as weather forecasting, drug discovery, and engineering design.
Anandkumar has made influential contributions to Machine learning, particularly through the development of tensor decomposition techniques for latent variable models and the invention of neural operators, which extend Deep learning to model multi-scale processes in scientific simulations. She has also worked on generalist AI agents using language models and has been a vocal advocate for diversity and inclusion in technology.
Early Life and Education
Anandkumar was born in Mysore, India, into a family of engineers and mathematicians. Her parents were both engineers, and her grandfather was a mathematician. Her great-great-grandfather was the Sanskrit scholar R. Shamasastry, who is known for discovering the Arthashastra. She began studying Bharatanatyam, a classical Indian dance form, at a young age and practiced it for many years.
She pursued an undergraduate degree in electrical engineering at the Indian Institute of Technology Madras, graduating in 2004. She then moved to the United States for graduate studies at Cornell University, where she earned a PhD in 2009 under the supervision of Lang Tong. Her doctoral thesis focused on scalable algorithms for distributed statistical inference. During her PhD, she was an IBM Fellow from 2008 to 2009 and worked on end-to-end service-level transactions in the networking group at IBM.
After completing her PhD, Anandkumar was a postdoctoral scholar at the Massachusetts Institute of Technology until 2010, working in the Stochastic Systems Group with Alan Willsky.
Academic Career and Tensor Methods
In 2010, Anandkumar joined the University of California, Irvine, as an assistant professor. At that time, the technology industry was at the beginning of the big data revolution, and she began working on tensor decompositions of latent variable models. Her work in this area provided new theoretical guarantees for learning probabilistic models from data, which became a cornerstone of her research.
In 2012, she was a visiting scientist at Microsoft Research in New England. In 2013, she received a National Science Foundation CAREER Award to investigate big data and social networks. She was promoted to associate professor with tenure at UC Irvine in 2016. Her research during this period focused on large-scale machine learning and high-dimensional statistics.
Industry Roles: Amazon Web Services and NVIDIA
From 2016 to 2018, Anandkumar served as a principal scientist at Amazon Web Services (AWS). At AWS, she worked on the Apache MXNet deep learning framework, introducing new functionality and developing multi-modal processing algorithms. She was involved in the launch of Amazon SageMaker, a platform that enables developers to build and deploy machine learning models. She also contributed to Amazon Rekognition, Amazon Lex, and Amazon Polly, which are AI services for image recognition, conversational interfaces, and text-to-speech, respectively.
In 2018, Anandkumar joined NVIDIA as director of machine learning research, where she opened a new core laboratory in artificial intelligence and machine learning in Santa Clara. She also became the Bren Professor of Computing and Mathematical Sciences at Caltech in the same year, a role she continues to hold. At NVIDIA, she has led research on AI for science, including neural operators and foundation models for scientific applications.
Neural Operators and AI for Science
Anandkumar has been a pioneer in developing AI algorithms for scientific discovery. She invented neural operators, which extend deep learning to model multi-scale processes in scientific domains. Unlike traditional neural networks that operate on discrete grids, neural operators learn mappings between function spaces, making them orders of magnitude faster than traditional simulations. This work has been applied to weather forecasting, drug discovery, and engineering design.
In 2021, neural operators were highlighted in Quanta Magazine as a featured achievement in mathematics and computer science. Anandkumar also developed AI-based high-resolution weather models and an AI-aided method for designing anti-infection medical catheters. In 2022, her work on genome-scale foundation models, which exhibit emergent behavior in predicting evolutionary dynamics and protein function, won the ACM Gordon Bell Special Prize for High Performance Computing-Based COVID-19 Research.
Generalist AI Agents and Language Models
Anandkumar has also contributed to early work on generalist AI agents using language models. Her research demonstrated how interactive in-context learning in language models can be used to construct actions in the form of program code to solve complex open-ended tasks in environments such as Minecraft and robotic reinforcement learning. This line of research explores how foundation models can simulate a life cycle of learning and adapt to new tasks interactively.
Advocacy and Public Engagement
Anandkumar has been an active campaigner for gender equality and against sexual harassment in technology and academia. She launched a petition to Timothy A. Gonsalves, then director of IIT Madras, to end gender segregation in the admissions process and to implement campus-wide systems to monitor sexual harassment, improve security, and increase alumni engagement. She has spoken openly about her own experiences of sexual harassment and called on Intel to stop using female acrobats at conference parties.
She was also one of the campaigners who successfully pushed to rename the Conference on Neural Information Processing Systems from 'NIPS' to NeurIPS. In 2018, she received a New York Times Good Tech Award for her advocacy and research.
Awards and Honors
Anandkumar has received numerous awards and honors throughout her career. In 2026, she was named a Time100 AI Innovator. In 2025, she received the Time100 Impact Award for using AI to accelerate scientific discovery and the IEEE Kiyo Tomiyasu Award for contributions to AI, including tensor methods and neural operators. In 2024, she won the Blavatnik Award for Young Scientists and was named a TED speaker. She also received the Distinguished Alumnus Award from IIT Madras in 2024.
In 2023, she was awarded a Guggenheim Fellowship in computer science, became a Schmidt Sciences AI 2050 Senior Fellow, and was elected as an AAAI Fellow. In 2022, she was named an ACM Fellow for contributions to tensor methods and neural operators. Her paper on neural operators also won an Outstanding Paper award at a major conference in 2022.
Legacy and Impact
Anandkumar's work has had a significant impact on both machine learning theory and its applications in science and engineering. Her tensor methods have provided a rigorous foundation for learning latent variable models, while her neural operators have opened new avenues for using AI in scientific simulation. Through her roles at Caltech and NVIDIA, she continues to shape the future of AI research, particularly in the emerging field of AI for science. Her advocacy for diversity and ethical practices in AI has also made her a prominent voice in the broader technology community.