AI will expand opportunity for Indian IT firms, but business models will have to evolve: Snowflake India MD

Artificial intelligence will expand the opportunity for Indian IT services companies, but it will also force them to rethink a business model built around billable hours and large workforces, according to Vijayant Rai, Managing Director of India at Snowflake.

Rai said customers were increasingly looking beyond AI experiments and pilots and demanding measurable business outcomes and returns on their technology investments. That shift could put pressure on traditional services models, but he does not see it shrinking the overall market for IT companies.

“We definitely believe that the market is expanding with AI,” Rai told CNBC-TV18. “We ourselves have seen that in our results, we’ve seen that in conversations with customers.”

The opportunity, he said, will increasingly lie in helping enterprises use AI across their existing technology and data estates, rather than simply deploying more people on technology projects.

AI will change how IT services are sold

The traditional Indian IT services model has largely been built around people, billable hours and scale. Rai expects AI to change that equation as customers become more focused on what they get from their technology spending.

“AI is also driving an outcome-based sort of paradigm with the end customer as well. They are looking at what kind of outcomes, what kind of ROI,” he said.

That does not necessarily mean fewer opportunities for IT services companies. Instead, Rai expects the nature of the work — and the way it is priced — to evolve.

“Services vendors will need to evolve as well with the new platforms which are in play. It’s an opportunity. It’s definitely an opportunity,” he said.

For Indian IT companies, that could mean a gradual move away from labour-intensive work towards higher-value services involving AI architecture, data engineering, domain expertise and implementation.

AI could unlock more enterprise work

Rai sees a particularly large opportunity in traditional enterprises, where significant amounts of data remain spread across legacy systems, SaaS applications and other systems of record.

Getting that data ready for AI still requires technical and domain expertise, even as AI tools automate parts of the process.

“That does require a lot of data engineering, for example. It does require a lot of understanding of that, even though we have tools which will sort of increasingly help and get better,” Rai said.

As companies begin deploying AI more widely, he expects system integrators and IT services companies to benefit from the expansion in workloads.

“As AI expands, it’ll start unlocking a lot of these systems of record in traditional enterprises that were previously being used. And that’s where we think that the system integrators do have a massive opportunity,” he said.

Indian IT companies have an advantage here because of their experience with large enterprises and their understanding of specific industries and technology environments.

The workforce advantage will have to change

AI also raises a bigger question for India’s IT services industry: how valuable will large workforces remain when software can increasingly automate parts of technology development?

Rai does not believe the answer is simply that fewer people will be needed.

Instead, he expects the composition of skills to change as AI opens up new areas of work. IT services companies will still need scale, but specialised technology and domain skills could become more important.

“They will have opportunities. Obviously, they’ll have to start getting more outcome-driven, and obviously, the narratives will change there,” Rai said.

He compared the shift with what happened in software-as-a-service businesses, where companies moved from seat-based models towards consumption-based measures such as storage and compute.

A similar change could happen in IT services as customers become more interested in outcomes than the number of people deployed on a project.

AI tools will not make technology skills redundant

Snowflake itself is developing tools that automate parts of software and data development. Its Cortex Code product, for instance, is designed to help users build data pipelines, analytics and AI applications faster.

But Rai disagrees with the idea that such tools will simply eliminate the need for technology professionals.

AI can make it easier for business users to prototype applications or build dashboards, but more complex tasks still require people who understand enterprise systems, data sources and business requirements.

“It has reduced the entry barrier, for sure. But at the same time, that also will increase the participation level with these technologies,” Rai said.

In other words, AI could allow more people within an organisation to work with technology while increasing the speed at which technology teams build and deploy applications.

From AI experiments to production

Rai’s comments come as Snowflake itself moves beyond its origins as a cloud data warehouse and positions itself as an AI Data Cloud.

The company said in its latest quarterly results that it had about 14,000 customers globally. Rai said around 13,600 of those customers were consuming AI features, compared with only a few thousand a year earlier.

The change reflects a broader shift in enterprise AI adoption. Companies that initially experimented with generative AI are increasingly looking at how to put AI into production and scale it across their organisations.

Rai said large enterprises were now building pipelines of AI projects, with some still in experimentation and others moving into production.

The challenge is increasingly about bringing these separate initiatives together.

Snowflake’s bet on the ‘agentic control plane’

Snowflake is positioning itself as an “agentic control plane” for enterprises, bringing together data, business context, AI models and information from existing systems.

Rai said context was particularly important because the same piece of data can have different meanings within different parts of an organisation. Snowflake is therefore focusing on semantic layers and business context alongside the underlying data.

The company also provides access to multiple frontier and open-weight AI models, allowing customers to select models based on their use cases and the outcomes they want to achieve.

That puts Snowflake in a competitive space alongside the hyperscalers on whose cloud platforms it operates. In India, Snowflake runs on AWS and Azure, while Google Cloud is also part of its global footprint.

Rai said Snowflake’s proposition was not to replace system integrators but to work with them.

Snowflake provides the platform and tools, while system integrators and consulting partners bring their understanding of the customer’s technology estate and industry.

India is moving at different speeds

AI adoption in India is not uniform. Digital-native companies, startups and unicorns are moving faster, while traditional enterprises are progressing through experimentation, pilots and production deployments at different rates.

Rai said many of the largest organisations Snowflake works with already have multiple AI projects under way.

The next phase will involve consolidating those efforts, particularly as enterprises find that data and AI models are spread across different parts of the organisation.

For Indian IT services companies, that creates both an opportunity and a challenge.

AI could expand the amount of technology work available, but customers are likely to demand more output from every rupee they spend. The companies best placed to benefit may therefore be those that can combine their existing domain and technology expertise with AI to deliver measurable business outcomes — rather than simply deploy more people.