How Tech Mahindra & Nvidia are Boosting AI Innovation

Companies worldwide are seeking to harness AI’s potential across various sectors and now, India, with its vast pool of tech capabilities and diverse linguistic landscape, is emerging as a key player in this field.

As AI technologies continue to evolve, there is an increasing focus on developing solutions tailored to specific regional and cultural contexts, particularly in the realm of natural language processing.

In a move to enhance India's AI capabilities, Tech Mahindra, a leading Indian information technology services and consulting company, has announced a strategic partnership with Nvidia, a renowned manufacturer of graphics processing units (GPUs) and AI hardware.

Together the companies aim to establish an AI Centre of Excellence (CoE) in India, focusing on developing advanced AI models tailored to the country's linguistic diversity, with a particular emphasis on Hindi and its numerous dialects.

Establishing the AI Centre of Excellence

The AI Centre of Excellence, powered by Nvidia's platforms, is designed to drive advancements in sovereign large language model (LLM) frameworks, agentic AI and physical AI.

The CoE leverages the Nvidia AI Enterprise software platform, which includes tools such as Nvidia NeMo for building language models and Nvidia RAPIDS for data analytics.

These advanced tools enable the development of customised, enterprise-grade AI applications capable of learning, reasoning and taking action—a concept known as agentic AI.

According to Tech Mahindra, the centre will also utilise the Nvidia Omniverse platform to develop connected industrial AI digital twins and physical AI applications.

Atul Soneja, Chief Operating Officer at Tech Mahindra, emphasised the significance of this collaboration, stating: "At Tech Mahindra, we are redefining the boundaries of AI innovation.

Collaborating with Nvidia, we are setting a new benchmark for enterprise-grade AI development by seamlessly integrating generative AI, industrial AI and sovereign large language models into the heart of global enterprises and industries".

Project Indus 2.0: advancing Hindi language models

A key solution emerging from this partnership is Project Indus 2.0, an advanced AI model focusing on Hindi and its numerous dialects, including Bhojpuri, Dogri and Maithili.

This project leverages Nvidia NeMo's capabilities to enhance conversations in Hindi across various sectors such as retail, banking, healthcare and citizen services in India.

The development of Indus 2.0 addresses the critical need for AI solutions that are culturally and linguistically relevant in India—a nation with over 19,500 languages or dialects spoken as mother tongues according to the Census of India.

By tailoring AI models to local languages, Tech Mahindra aims to provide more nuanced and effective solutions that resonate with India's diverse population.

John Fanelli, Vice President of Enterprise Software at Nvidia, commented on the collaboration, saying: "Built with Nvidia technology, Tech Mahindra's Center of Excellence will accelerate the development and adoption of sovereign AI LLMs and applications tailored for India's diverse industries and linguistic landscape.

“Our collaboration with Tech Mahindra is helping create a foundation for AI-driven innovation in the region and around the world".

This partnership between Tech Mahindra and Nvidia is part of a larger trend of collaboration between global technology leaders and Indian IT firms.

Major players like TCS, Infosys and Wipro are developing AI solutions using Nvidia's technology stack.

TCS launched a Gen AI solution for manufacturing for instance, while Infosys is using Nvidia 

AI enterprise for its Topaz offerings

These initiatives are expected to create new job opportunities, with Goldman Sachs forecasting AI's potential to automate up to 20% of work tasks in emerging economies.

At the Nvidia AI Summit in Mumbai, it was also revealed that companies like Ola Electric are working with Nvidia on initiatives such as using Omniverse for Digital Twin platforms.

Nvidia additionally released Nemotron-4-Mini-Hindi-4B, a small Hindi language model, which Tech Mahindra uses alongside NeMo to develop Indus 2.0.

As Atul concludes: "This integration provides customers with enhanced observability, granular control and optimised performance for their AI workloads across the entire stack, from the hardware acceleration layer to the application interface".

******

Make sure you check out the latest edition of AI Magazine and also sign up to our global conference series - Tech & AI LIVE 2024

******

AI Magazine is a BizClik brand