Tech Mahindra Ltd. has announced the establishment of a centre of excellence powered by Nvidia platforms to drive advancements in sovereign large language model frameworks, agentic artificial intelligence and physical AI.
Based on the Tech Mahindra Optimized Framework, the CoE leverages the Nvidia AI Enterprise software platform, including NeMo, NIM microservices and Rapids, to offer customised, enterprise-grade agentic AI applications. Agentic AI helps improve productivity by enabling AI applications to learn, reason and take action.
The CoE also uses the Nvidia Omniverse platform to develop connected industrial AI digital twins and physical AI applications across sectors, including manufacturing, automotive, telecommunications, healthcare and banking, financial services and insurance.
Leveraging the capabilities of the CoE, Tech Mahindra has also developed Project Indus 2.0, an advanced AI model powered by NeMo based on Hindi, Bhojpuri, Dogri and Maithili. Project Indus 2.0 caters to sectors, such as retail, banking, healthcare and citizen services in India. In the future, Indus 2.0 aims to include agentic workflows and support multiple dialects.
"Collaborating with Nvidia, we are setting a new benchmark for enterprise-grade AI development by seamlessly integrating gen AI, industrial AI and sovereign large language models into the heart of global enterprises and industries," Atul Soneja, chief operating officer of Tech Mahindra, said.
Tech Mahindra will also leverage the new Nvidia NIM Agent Blueprint for customer service to help call centre clients build custom AI virtual assistants that can aid human agents in rapidly resolving issues.
"Built with Nvidia technology, Tech Mahindra's Centre of Excellence will accelerate the development and adoption of sovereign AI LLMs and applications tailored for India’s diverse industries and linguistic landscape," John Fanelli, vice president of enterprise software at Nvidia, said.
The CoE is located within Tech Mahindra's Makers Lab in Pune and Hyderabad.