Dream It, Build It, Ship It: India's Multilingual Leap in Agentic AI India's readiness is further underscored by its rapid rise in AI talent and multilingual NLP capabilities, enabling developers and non-developers, even in remote regions, to build applications in regional languages.

India is emerging as a dominant force in the global Agentic AI landscape. According to NASSCOM's recent Generative AI Tracker, Indian GenAI startups have surged 3.7 times in number to over 890 by H1 CY2025, with approximately 83 per cent focused on vertical, application-oriented solutions.

Nearly 25 per cent of GenAI startup funding in the second half of FY25 was channeled to Agentic AI, emphasizing workflow automation, legal, and ITSM solutions.

Ali Rey, Group Head Technology, Emirates NBD spoke about the growing duality of Agentic AI and empowering employees through AI tools and said at a Nasscom event held in Bengaluru, "The agent is coming around for good and for bad, and we need to make sure we leverage it for good use cases."

Rey said that Emirates has gone all in and a large portion of the corporation has open AI or co-pilot within the office applications.

"We have GitHub with co-pilot. We've seen that they're improving from a productivity point of view 20, 30, 40 per cent. We run models for the GPUs on-premise… we really want to roll that AI across the bank as much as possible and take advantage of it," added Rey.

Ali Rey, Group Head Technology, Emirates NBD spoke about the growing duality of Agentic AI and empowering employees through AI tools and said, "The agent's coming around for good and for bad, and we need to make sure we leverage it for good use cases."

Rey said that Emirates has gone all in and a large portion of the corporation has open AI or co-pilot within the office applications.

"We have GitHub with co-pilot. We've seen that they're improving from a productivity point of view, 20, 30, 40 per cent. We run models for the GPUs on-premise… we really want to roll that AI across the bank as much as possible and take advantage of it," added Rey.

The Enterprise Experiments with AI Agents – 2025 Global Trends report reveals that 88 per cent of enterprises plan to allocate specific budgets to test and build AI agents, with over 60 per cent already progressing from pilot to production stages.

Another NASSCOM insight indicates enterprises are expected to spend 3-4 times more on AI agents in 2025, redirecting investments from traditional AI toward interactive, autonomous agent-based systems.

Ankit Bose, Head of AI, Nasscom, said that the benefit of AI to businesses and service companies is separate.

"But we did a study last year, and what we realised was that developer satisfaction effectiveness has increased way above right. All the developers are using AI in some form. More and more enterprises are opening up for that," said Bose.

On the enterprise side, 88 per cent of organizations globally now have dedicated budgets to experiment with AI agents, and over 60 per cent are moving from pilots to production, reflecting the "prompt, generate, verify" software lifecycle.

Karan M V, International Developer Relations Director at GitHub, on Agentic AI redefining the software lifecycle and India's cultural readiness for agentic interactions, said, "This is what's happening, redefining how software itself is built. This is evolving into a prompt, generate, verify. The new SDLC. We are used to working with agents, even in the physical world here in India. This is actually a part of our DNA of working with agents."

India's readiness is further underscored by its rapid rise in AI talent and multilingual NLP capabilities, enabling developers and non-developers, even in remote regions, to build applications in regional languages.

"A student, even remote, can start off their journey by prompting in their own natural language, whether it be Kannada or Hindi. Imagine building entire apps only using natural language and ship it," added Karan.