Sashikumar Ganesan, professor, Indian Institute of Science (IISC), Bengaluru, his journey into AI research wasn’t as straightforward. Trained in computational mathematics for about two decades, Ganesan’s entry into AI research happened in 2020—during the Covid-19 outbreak—when he was roped in, along with few colleagues, into building a model to predict the pandemic’s trajectory. “I was a core computational mathematics person. But since Covid, I have gone into Python, and machine learning neural networks. I want to bring this into science and engineering and that is how the research started,” Ganesan said.
The AI for Research and Engineering eXcellence (AiREX) – a lab Ganesan heads in IISc—is focused on using AI to tackle scientific and engineering issues across engineering sectors such as manufacturing. Ganesan and IISc are not alone.
Academics across the country are increasingly getting immersed in AI research, hoping to uncover its potential. Few hundred kilometres from IISc, Manjunath K N, associate professor in computer science and engineering at Manipal Institute of Technology in Manipal, Karnataka, along with a team of experts are deep at work building “coronary artery disease diagnosis and prediction systems using AI”. After a Rs 81-lakh funding for proof of concept (PoC), the project—in its second phase—secured Rs 300 crore in funding from the education ministry, and `30 crore from industry partners. “Our work focuses on developing the deep learning base model for improvising the cardiology workflow,” he said.
This project initiated by the government is primarily focussed on creating affordable healthcare services, impacting those at the grassroot level and not just the rich. This is being done using the AI platform in collaboration with researchers from other leading institutes in the country. All the way in New Delhi, Manan Suri, associate professor, and principal investigator, neuromorphic hardware research group at Indian Institute of Technology (IIT) Delhi and his team are working on designing less power-consuming hardware for sustainable compute—a hot area currently given how expensive computing is. The team is focusing on reducing energy consumption, improving computation speed, and on ways to make the chip more compact.
“Most of our effort is directed towards finding a sweet spot or to extract the best while optimising one or more of these metrics,” said Suri.
In another department in IIT-Delhi, Manabendra Saharia, an associate faculty at the Yardi School of Artificial Intelligence, is engaged in monitoring and mitigating natural hazards such as floods, droughts, and landslides using AI. Saharia is collaborating with Indian space agency ISRO, Ministry of Earth Sciences, and international partners to develop systems that can map and forecast disasters in advance. “The goal is to make hazard forecasting more scalable, adaptive, and useful for decision-makers,” he said.
While some of the work by institutes has garnered industry attention, others are still confined to being just research projects. Unlike in the US, the private sector in India lags in participating in research and development. This, experts say, needs to pick up. V. Ramgopal Rao, group vice-chancellor at BITS Pilani said India is not doing enough when it comes to contribution of institutes in driving advancements in new-age technology like AI. “China’s Deepseek and the West’s OpenAI show what can be achieved when academia, government, and industry align with strategic clarity and serious investment,” he said. “Indian academia is talented but resource-constrained, with limited access to compute, data, and translational pathways,” said Rao, previously a director at IIT Delhi and chair professor for nanoelectronics at both IIT Bombay and IIT Delhi.
To be sure, while the government is investing resources to advance AI research, the gestation period of initiatives is still high.
AI Research in Academia
IISc BENGALURU
Sashikumar Ganesan’s team at IISc is collaborating with IBM and engineering companies to do science with help of LLMs. While LLMs can discuss scientific topics, they cannot yet ‘do science and engineering’ in the way researchers and engineers need. To overcome this limitation, Ganesan and his startup team are working on enabling foundational models to ‘do science’ by integrating computational physics, domain knowledge & equation-solving capabilities. This research aims to build models that go beyond text generation to deliver scientifically validated outputs
IIT DELHI
Manabendra Saharia at IIT Delhi is involved in developing systems that can map and forecast disasters in advance. This is in collaboration with Isro, Ministry of Earth Sciences and global partners. The most recently developed AI model is called DeepSARFlood that automatically maps floods through satellite radar data that penetrates clouds and works round the clock. The project is funded by the IndiaAI mission and is being extended to monitor landslides and damaged infrastructure.
DELHI TECHNOLOGICAL UNIVERSITY
IT department head Dinesh K Vishwakarma and his team are working to create AI models for online fake news and hate speech detection, trained on real-life incidents. They are also working on AI model vulnerability identification using ‘noise’ attacks and measuring success ratios for accuracy. Researchers also collaborate with agriculture scientists on smart agriculture and crop disease detection AI. Vishwakarma worked on AI-powered human activity recognition for smart homes, with the aim of applying this to senior care.
IIT HYDERABAD
Maunendra Sankar Desarkar, an associate professor in IIT Hyderabad’s Department of AI is focusing on Natural Language Processing or NLP or foundational language models or Large Language Models (LLMs). Most of the LLMs that people are using now are not from India and have very little Indian content, he says. To find a solution, Desarkar is working on a model called BharatGen, which is an initiative supported and funded by the government of India. The institute also has a Technology Innovation Hub for Autonomous Navigation (TIHAN), funded by the Department of Science and Technology under the National Mission on Interdisciplinary Cyber-Physical Systems. TIHAN has India’s first Autonomous Navigation Testbed (Aerial & terrestrial) facility, according toDesarkar. He along with other faculty members at the AI department are also working on use of AI for specific domains like healthcare, agriculture, astrophysics, process optimization, networks, etc. “IIT Hyderabad has already developed a driverless campus shuttle that offers mobility services across the institute campus,” he said. A passenger-carrying drone with 150 kg payload capacity is going to be tested soon.
Annapurna Roy in New Delhi contributed to this story.
#Pahalgam Terrorist Attack
Pakistan's economy has much more to lose than India's due to the ongoing tensions, warns Moody's Ratings
The day Pakistan got the power to poke India
FM Sitharaman meets ADB chief and Italian FM, discusses economic issues; no mention of Pakistan
Academics across the country are increasingly getting immersed in AI research, hoping to uncover its potential. Few hundred kilometres from IISc, Manjunath K N, associate professor in computer science and engineering at Manipal Institute of Technology in Manipal, Karnataka, along with a team of experts are deep at work building “coronary artery disease diagnosis and prediction systems using AI”. After a Rs 81-lakh funding for proof of concept (PoC), the project—in its second phase—secured Rs 300 crore in funding from the education ministry, and `30 crore from industry partners. “Our work focuses on developing the deep learning base model for improvising the cardiology workflow,” he said.
This project initiated by the government is primarily focussed on creating affordable healthcare services, impacting those at the grassroot level and not just the rich. This is being done using the AI platform in collaboration with researchers from other leading institutes in the country. All the way in New Delhi, Manan Suri, associate professor, and principal investigator, neuromorphic hardware research group at Indian Institute of Technology (IIT) Delhi and his team are working on designing less power-consuming hardware for sustainable compute—a hot area currently given how expensive computing is. The team is focusing on reducing energy consumption, improving computation speed, and on ways to make the chip more compact.
“Most of our effort is directed towards finding a sweet spot or to extract the best while optimising one or more of these metrics,” said Suri.
In another department in IIT-Delhi, Manabendra Saharia, an associate faculty at the Yardi School of Artificial Intelligence, is engaged in monitoring and mitigating natural hazards such as floods, droughts, and landslides using AI. Saharia is collaborating with Indian space agency ISRO, Ministry of Earth Sciences, and international partners to develop systems that can map and forecast disasters in advance. “The goal is to make hazard forecasting more scalable, adaptive, and useful for decision-makers,” he said.
While some of the work by institutes has garnered industry attention, others are still confined to being just research projects. Unlike in the US, the private sector in India lags in participating in research and development. This, experts say, needs to pick up. V. Ramgopal Rao, group vice-chancellor at BITS Pilani said India is not doing enough when it comes to contribution of institutes in driving advancements in new-age technology like AI. “China’s Deepseek and the West’s OpenAI show what can be achieved when academia, government, and industry align with strategic clarity and serious investment,” he said. “Indian academia is talented but resource-constrained, with limited access to compute, data, and translational pathways,” said Rao, previously a director at IIT Delhi and chair professor for nanoelectronics at both IIT Bombay and IIT Delhi.
To be sure, while the government is investing resources to advance AI research, the gestation period of initiatives is still high.
AI Research in Academia
IISc BENGALURU
Sashikumar Ganesan’s team at IISc is collaborating with IBM and engineering companies to do science with help of LLMs. While LLMs can discuss scientific topics, they cannot yet ‘do science and engineering’ in the way researchers and engineers need. To overcome this limitation, Ganesan and his startup team are working on enabling foundational models to ‘do science’ by integrating computational physics, domain knowledge & equation-solving capabilities. This research aims to build models that go beyond text generation to deliver scientifically validated outputs
IIT DELHI
Manabendra Saharia at IIT Delhi is involved in developing systems that can map and forecast disasters in advance. This is in collaboration with Isro, Ministry of Earth Sciences and global partners. The most recently developed AI model is called DeepSARFlood that automatically maps floods through satellite radar data that penetrates clouds and works round the clock. The project is funded by the IndiaAI mission and is being extended to monitor landslides and damaged infrastructure.
DELHI TECHNOLOGICAL UNIVERSITY
IT department head Dinesh K Vishwakarma and his team are working to create AI models for online fake news and hate speech detection, trained on real-life incidents. They are also working on AI model vulnerability identification using ‘noise’ attacks and measuring success ratios for accuracy. Researchers also collaborate with agriculture scientists on smart agriculture and crop disease detection AI. Vishwakarma worked on AI-powered human activity recognition for smart homes, with the aim of applying this to senior care.
IIT HYDERABAD
Maunendra Sankar Desarkar, an associate professor in IIT Hyderabad’s Department of AI is focusing on Natural Language Processing or NLP or foundational language models or Large Language Models (LLMs). Most of the LLMs that people are using now are not from India and have very little Indian content, he says. To find a solution, Desarkar is working on a model called BharatGen, which is an initiative supported and funded by the government of India. The institute also has a Technology Innovation Hub for Autonomous Navigation (TIHAN), funded by the Department of Science and Technology under the National Mission on Interdisciplinary Cyber-Physical Systems. TIHAN has India’s first Autonomous Navigation Testbed (Aerial & terrestrial) facility, according toDesarkar. He along with other faculty members at the AI department are also working on use of AI for specific domains like healthcare, agriculture, astrophysics, process optimization, networks, etc. “IIT Hyderabad has already developed a driverless campus shuttle that offers mobility services across the institute campus,” he said. A passenger-carrying drone with 150 kg payload capacity is going to be tested soon.
Annapurna Roy in New Delhi contributed to this story.
