Indian-American researcher Vidya Chhabria has been named the recipient of the first Google ML and Systems Junior Faculty Awards.
Assistant Professor of Electrical Engineering in the Department of Electrical, Computer and Energy Engineering at Arizona State University is in Computer Aided Design (CAD) for very large-scale integration, or VLSI systems.
According to the university's release, she is one of more than 50 assistant professors from 27 US universities that have been awarded for the award by a prominent group of Google engineers and researchers.
Chhabria emphasizes Google's commitment to working with academia and its perception of research that utilizes machine learning for hardware design.
“Being recognized by Google through this Junior Facility Award is extremely rewarding not just for me, but for our entire group,” says Chhabria. “Our group focuses on developing specialized software, electronic design automation tools, that help design computer chips behind everything, from smartphones to data centers.”
Her research group develops electronic design automation tools, specialized software that helps design computer chips for use in everything from smartphones to data centers.
Read: Startup PartCL, a revolutionary chip design with AI-powered speed (
“Chip designs are complex, time-consuming and resource-intensive, and AI shows great potential in addressing the scale, automation and optimization challenges of this space,” she says.
In addition to approval of the award, Chhabria receives $100,000 with unlimited funding.
She supports research projects and highlights the importance of this resource in moving faster and increasing scale when using AI to automate computer chip designs.
“Training AI for chip design is particularly difficult due to the lack of open source. Industrial-scale chip design makes creating and testing AI models difficult,” she says. “The award helps bridge that gap by creating opportunities for collaboration with Google. Mentorship and access to industry perspectives, and perhaps computational resources, will keep the work industry relevant and advance chip design AI.”
As chips become more complex, tools struggle with scalability and optimization, as they are traditional electronic design automation, or EDA. By advancing AI-driven EDA tools, her group at ASU is working on next-generation chip design technology. This is a tool that allows you to use AI to build the hardware needed for future AI applications.
“This work places ASU as a leader in microelectronics, not only as a leader in EDA, but in particular as a leader in EDA, which is essential to shaping the future of the semiconductor industry,” she says.
Chhabria is excited to use the Big Language Model (LLM) as an intelligent “agent” for chip design.
Agents can learn to perform design tasks automatically and improve efficiency and optimization while lowering non-professional entry barriers. For example, in physical design, the process is extremely complex and requires a careful balance of speed, power, area and manufacturing possibilities, as abstract circuits are at the stage of chip design that transforms into actual chip layouts of silicon.
Read: Kaushik Sengupta leads a $10 million project to create advanced semiconductors (
“Developing autonomous AI agents for a variety of challenges in physical design could be transformative,” she says.
Chhabria emphasizes that building such agents requires a large amount of high quality, labeled training data, which is currently rare in chip designs. Therefore, another great thrust in her work is the creation of synthetic data using the generated AI.
Synthetic data generation fills the key gaps in open source designs and ensures that academic research remains relevant and competitive.
“By creating realistic, large datasets that do not exist in real life, we can train AI models more effectively and make progress even in regions where access to industrial data is limited,” she adds. “Together, these two directions, AI agents for design and data generation AI will help redefine what is possible in the future of EDA.”
For society, the task of generating faster and more efficient chip designs leads to cheaper, environmentally friendly, faster available technologies.
Chhabria graduated from the University of Minnesota in 2022 and 2018 with a PhD and Masters degree.
