Indian IT’s next phase of growth will not come from adding thousands of engineers to delivery teams, but from fundamentally reshaping how technology services are built, sold and scaled in an AI-first era.
That was the central message from industry leaders at an Umagine TN 2026 panel discussion on “Unlocking the Next Wave of IT Growth: AI, Platforms & Digital Innovation in Tamil Nadu.”
Srinivasan Panchapakesan, senior corporate vice president and global delivery head of digital and software at Hexaware Technologies, framed the shift in stark economic terms. The traditional services model, where hiring 1,500–2,000 people translated into incremental revenue, was being dismantled by AI-led delivery, he said.
He argued that the industry was facing an unprecedented opportunity rather than an existential threat. Jobs would not disappear, he said, but the nature of work would change decisively. “A person who knows AI is going to be more challenging than a person who doesn’t know AI.”
The next growth cycle, according to Panchapakesan, will be driven by “copilot-driven, collaborative, AI-supported models,” rather than pure headcount expansion.
Regional Data Starvation
If AI is the growth engine, data, especially local, contextual data, is the fuel. Ganesh Sankaralingam, director and delivery head at LatentView Analytics, argued that Tamil Nadu and other states’ biggest AI opportunity lies in their linguistic and sectoral diversity, but warned that India is dangerously underprepared on data creation in regional languages.
“To build Tamil large language models, we need a corpus of data in Tamil, which is a mandate,” he said. Yet, he pointed out, most people do not type in Tamil, even on messaging platforms, starving AI systems of training data.
Sankaralingam traced the problem back to the education system, suggesting that digital literacy in regional languages must begin in primary school if Tamil and other Indian languages are to thrive in an AI-driven world.
The market reality, he said, is overwhelmingly regional. He cited practical use cases such as analysing emergency call transcripts to surface public safety trends, problems that can only be solved when local data exists.
Playing to Local Strengths
While data enables intelligence, platforms are emerging as the dominant economic structure through which AI will be monetised. Janardhan Santhanam, CIO at Tata Consultancy Services, said the industry’s evolution, from infrastructure-as-a-service to software, business process services and now agentic AI, has placed platforms at the centre of value creation.
Tamil Nadu, he argued, is unusually well-positioned as a test bed because of its mix of manufacturing, automotive and textile industries.
Santhanam pointed to digital twin platforms in process industries, where real-time data and predictive algorithms can simulate supply chains and production systems tailored to Indian conditions. In automotive manufacturing, he highlighted AI platforms that combine computer vision, generative and agentic AI with robotics, designed for shop floors that look very different from global warehouses.
In textiles and crafts, where Tamil Nadu accounts for 28% of total employment in the sector, generative design platforms enable consumers to co-create products while preserving artisan employment. He cited a TCS initiative that blends AI, IoT and immersive technologies to translate digital designs into handwoven output. The common thread, he said, is that platforms built for Indian realities can become globally relevant products.
Talent Depth Deficiency
Building such platforms for global markets, however, comes with structural challenges. Raj Radhakrishnan, vice-president of engineering at NielsenIQ, said success requires engineering at scale, not just building strong code.
Global platforms must meet international standards across user experience, security and reliability, while remaining locally adaptable. Talent, he added, remains a bottleneck, not in quantity, but in depth. Sustained investment in R&D, cybersecurity, MLOps and deep tech capabilities will determine whether platforms built in Tamil Nadu can compete globally.
Compliance is another constraint. Platforms serving international clients must meet regulations such as GDPR and SOC 2 while adhering to Indian laws.
Radhakrishnan said Tamil Nadu’s data centre policies and digital infrastructure provide a base, but long-term competitiveness will also depend on green, globally resilient infrastructure.
Underpinning all of this, he argued, is culture. “Collaboration is non-negotiable,” he said, calling for tighter alignment between academia, industry and government to move from participation to leadership in the global IT landscape.
Degrees Not Enough for AI
As the delivery model and technology stack evolve, talent remains the most immediate pressure point. Panchapakesan was blunt in distinguishing education from employability.
Degrees alone, he said, do not prepare candidates for AI-led work. While governments and industry are investing in labs, hackathons and internships, students must actively seize those opportunities, focusing on practice, problem-solving, and continuous learning.
Santhanam extended the argument to mid-level professionals, warning that AI and automation will rapidly erode traditional coordination and people-management roles. Large firms like TCS, he said, are already redeploying mid-level managers into specialised technical roles and certifying them in cloud, enterprise platforms and AI. He welcomed recent reskilling initiatives involving IIT Madras, Google, and the Tamil Nadu government, but stressed that responsibility ultimately lies with individuals.
Sankaralingam offered a pragmatic lens for employability: domain knowledge combined with AI skills. Candidates, he said, routinely underestimate the importance of understanding industry-specific problems, whether in financial services, automotive, or consumer analytics.
IP: The Golden Goose
The speakers stressed that Indian IT’s shift toward AI-led platforms will only translate into durable growth if it is anchored in intellectual property creation and reuse, rather than one-off deliveries.
He said that creating globally relevant IP also depends on embedding design thinking, privacy-first architectures and secure data practices early, starting at the school and college level.
Santhanam framed IP in two parts: creation and reuse.
On creation, he pointed to large-scale AI hackathons as a powerful mechanism to drive grassroots innovation and surface ideas that can later be industrialised into platforms. He cited TCS’s internal AI hackathon, which saw participation from over 2,80,000 employees, and argued that similar ecosystem-wide initiatives could significantly expand the pool of reusable IP.
On reuse, Santhanam said not every patent or innovation becomes a product, but unused IP still carries economic value if it can be discovered, licensed or reused. He suggested that a structured IP marketplace, supported by clearer visibility into existing patents and data assets, could accelerate this process.
Panchapakesan said IP-backed platforms have already become central to service delivery. However, he emphasised that IP creation is ultimately an outcome of attitude rather than numbers, continuous learning, disciplined planning and thoughtful execution.
The post How Indian IT’s AI-led Growth Could Hinge on Regional Data, Talent appeared first on Analytics India Magazine.
