Doctoral student Vidya Chhabria receives Women in Technology scholarship by Cadence

Doctoral student Vidya Chhabria is a recipient of Cadence’s Women in Technology scholarship. The scholarship program seeks to foster inclusion and encourage diverse backgrounds, experiences, and ideas in keeping with the company’s goal to support and expand diversity in technology within academia and in the workplace.

Vidya’s research focuses on electronic design automation (EDA) which helps in the systematic, efficient, and rapid design of complex electronic circuits. These circuits have billions of microscopic components that are tightly packed into small packages. The increased complexity of compact electronic systems such as implantable medical devices and cell phones have led to increased chip power densities. This can affect performance and cause heating issues which impact battery life, and lead to device failure. Vidya’s research primarily addresses this challenge. Her work involves developing novel algorithms that leverage machine learning (ML) techniques to automatically design power delivery networks and analyze temperature, performing both tasks across the entire chip. 

Vidya earned her bachelor’s degree in India. While an undergraduate student, her senior design project in digital logic design proved to be a turning point: the experience cemented her interest in the field. As a graduate student at the University, she has been working under the guidance of Prof. Sachin Sapatnekar, contributing towards research in the field. For her, the field combines her enthusiasm for algorithms and hardware design, which enable the development of complex systems with diverse applications. The interdisciplinary nature of EDA, and its capacity to build advanced systems with extensive applications motivate her interest in the field. 

EDA tools typically design chips using heuristic methods which have been rather effective. With the help of machine learning (ML), chip designers can now leverage decades worth of designs by turning the data into valuable insight. ML-based EDA tools minimize errors in new designs, and reduce costs and turnaround times.

Chip design costs have risen exponentially over the years because of two key reasons: the prohibitive costs of EDA software, and the cost of chip fabrication. Vidya’s research has the potential to make a valuable contribution towards reducing costs. She is working on building open-source software by contributing to the OpenROAD project (Foundations and Realization of Open Accessible Design), an effort that involves over 30 researchers across four universities that aims to create a public-domain EDA toolchain. She is also working towards establishing a new machine learning (ML) paradigm that develops novel software to analyze the impact of the high power densities on power delivery network design, and temperature for advanced chips.