Vidushi Sharma (Ph.D.) is a Computational Material Scientist with domain expertise in the fields of energy storage and nanomedicine. Currently, she is a Staff Research Scientist at the IBM Almaden Research Center, San Jose, California, where she performs incisive research in the field of Energy Storage, based on next generation themes such as artificial intelligence driven discovery of new battery materials (electrolytes and electrodes), development of new artificial intelligence algorithms for complex materials (foundation models for mixtures, heterostructures and composites), and application of quantum computing in material discovery. Dr. Sharma completed her Ph.D. in Mechanical Engineering (with focus on computational material science) at New Jersey Institute of Technology (NJIT), USA. She has authored 30+ research papers, and several of her works are featured in major media outlets.
Award
IBM Research has a dedicated focus on Accelerated Discovery of Sustainable Materials, and one facet of that work is an active research effort in energy storage, particularly batteries. The work is primarily centered at IBM Research – Almaden in San Jose, CA, in collaboration with IBM’s global research team. We combine both experimental and cutting-edge computational technologies including artificial intelligence to explore, develop, and validate new, more sustainable battery materials and chemistries capable of supporting clean transportation and renewable energy infrastructure.
We are seeking a highly motivated and talented undergraduate or graduate student to join our research team focused on developing innovative battery technologies using generative AI. This internship will provide you with a unique opportunity to contribute to cutting-edge research and gain hands-on experience in the field of AI and materials science. The intern will work on a research project with the objective to fine-tune a Large Language Models (such as LLaMA or equivalent) for predicting suitable electrolytes for a given anode-cathode system in battery applications. By leveraging the power of LLMs, we aim to expedite the electrolyte selection process and enhance the performance and safety of battery technologies. The project would constitute data collection on anode-cathode combinations and their compatible electrolytes, fine-tuning the appropriate LLM on the curated dataset using techniques like transfer learning and prompt engineering, and developing the prediction framework. The intern will collaboratively work with lab scientists to experimentally validate and demonstrate the successful application of LLMs in energy storage applications.
From June 16 to September 5, 2025 (adjustable at the discretion of the organisation)