Indian IT’s Agent Era has Begun

All Articles on Technology

The Indian IT services industry is undergoing a transformation so profound that it might soon become unrecognisable. This goes far beyond a routine shift to the cloud or buzzword-driven push into so-called “digital transformation”. This time, it’s all about agents. Specifically, autonomous AI agents. The shift is evident not just in earnings calls and strategy decks, but also in headcount, delivery models, organisational charts and even revenue structure. Q1 FY26 makes one thing clear: Indian IT firms are no longer merely preparing for AI; they are already deploying it. Moreover, they are doing so at a scale that few outside the industry had anticipated. TCS Shrinks to Grow TCS, the ₹12 lakh crore behemoth, reported ₹63,437 crore in revenue this quarter, a modest 1.3% YoY increase. But the real headline wasn’t the growth. It was the reduction of over 12,000 employees, the steepest drop in headcount in recent quarters.  While the management attributed it to “efficiency improvements”, the writing is on the wall: agentic systems are beginning to replace roles that lack scalability. TCS is also reorganising its teams around AI-led practices, platform-based offerings and reusable agents designed to plug into client systems with minimal friction. The company isn’t just offering AI services; it’s embedding AI into the services model itself. Yet, even as the shift gathers pace, TCS chief innovation officer Bala Prasad Peddigari sounds a note of caution. In an exclusive interaction with AIM, he pointed out that most agentic systems today still struggle with accuracy, domain-specific fine-tuning and contextual awareness, making enterprise-grade deployment a work in progress, not a done deal. TCS, however, is taking a long-view approach by building 360-degree ecosystem partnerships and deeply integrating AI into its delivery backbone, rather than chasing flashy one-off deployments. Infosys Keeps It Quiet and Real If TCS is in the midst of restructuring, Infosys is already shipping. The company has deployed over 300 AI agents across internal operations and client-facing services. Their AI stack, Topaz, blends proprietary models, open weights and third-party integrations. With ₹41,206 crore in revenue, 3.8% growth, and $3.8 billion in large deals, Infosys didn’t just speak of AI; it executed. Infosys CEO Salil Parekh called it “industrial-grade AI”, the kind that replaces 10 slide decks with one line of working code. Infosys also trained 95% of its delivery staff in GenAI basics and use cases, not in labs, but on live projects. [Must Watch] Check out our exclusive coverage of Infosys’ latest earnings—only on Front Page. Wipro’s Growing Pains Wipro’s revenue declined 2.3% YoY to ₹22,530 crore, its weakest performance in recent memory. However, its GenAI bets are surprisingly aggressive. The company secured $5 billion in deal bookings this quarter, alongside building and deploying over 200 AI agents, and reported a GenAI-anchored deal pipeline worth $2.7 billion, a staggering 131% increase compared to the same period last year. Notably, every one of the 63 large deals Wipro signed included GenAI use cases, from smart cloud agents to AI observability layers. The company reorganised its global business lines to consolidate AI, analytics and infrastructure under a unified tech services vertical. It also trained over 87,000 employees in GenAI tools, one of the highest among peers. Since taking over as CEO in April 2024, CEO Srini Pallia has led Wipro through a critical pivot, away from disjointed digital promises towards cohesive AI delivery. Pallia’s focus is clear: turn Wipro into a builder’s company, not just a billing partner. “Every major deal now has AI at its core,” he declared recently, underscoring Wipro’s move from experimentation to scaled execution. HCLTech Bets on OpenAI, Not Just Azure While others access GPT through Microsoft, HCLTech is working directly with OpenAI, embedding models into its enterprise offerings and federated agent stacks. The company reported ₹30,349 crore in revenue, marking a 3.7% YoY growth and has already trained 1,27,000 employees in generative AI. However, HCL’s agent focus is different; it’s aimed at productising automation. They are rolling out a series of AI-augmented tools across ITSM, infra and cybersecurity, wrapped around OpenAI’s APIs and their own agent layer. While profits dipped 10% this quarter, the long-term play is clear: turn every repetitive task into a GPT-powered agent. Tech Mahindra’s Surprise Comeback After multiple muted quarters, Tech Mahindra posted a 34% profit jump this quarter. The numbers? ₹13,351 crore in revenue (up 2.7%), over 200 agents in production and 77,000 employees trained in GenAI. Much of this is powered by KOGO, Tech Mahindra’s in-house orchestration platform, which builds compliant and observable agents for enterprise needs. Their focus is on industrial agent frameworks, compliant by design, reusable across verticals and optimised for hybrid infrastructure. Mid-Sized IT is Playing Offence While the Big Five transition cautiously, mid-tier IT firms are going all in. These firms aren’t experimenting with GenAI, they are shipping it into BFSI, manufacturing, retail and travel. Persistent Systems, for example, is leading ERP migration projects powered entirely by agents. LTIMindtree is building modular orchestration layers for enterprise clients, using open-source LLMs and custom agent frameworks. Smaller size = faster loops = quicker agent integration. The result? Higher growth, lower bloat. Where’s the AI Revenue? Everyone’s building agents. But how many are billing for them? TCS? No revenue disclosure. Infosys? No AI-specific number. HCLTech? Deal wins down from 12 last quarter to nine now. Accenture? $1.5 billion in GenAI bookings just this quarter. Indian IT’s agent push is impressive. However, clarity on monetisation is still missing. That could become a problem, especially when clients start demanding ROI from their AI pilots. Indian IT’s Future: From Pyramid to Agent Grid The traditional IT delivery model was clear: bench, pyramid, utilisation and billing. However, agents don’t need benches. Or layers. Or even human managers. They need orchestration, observability, fine-tuning and continuous feedback loops. This quarter signals the beginning of that rewrite: Fewer FTEs, more autonomous processes Platform-led growth, not just people-led Agents as revenue units, not just support tools This isn’t just a tech transition. It’s an org shift, a mindset reset, and perhaps, Indian IT’s most defining transformation since the Y2K era. That’s all for this week. If you found this newsletter insightful, share it with a friend, a colleague, or that one person who still thinks Indian IT can’t build AI beyond PoCs. And if you haven’t subscribed to our AIM Tv channel yet, now’s a good time. We break down the world of tech and AI in real time—crisp, clear and always ahead of the curve. While you are at it, here’s a quick look at some of the top stories of the week. Cognizant is betting $1 billion in ‘vibe coding’ to make AI write 50% of all code, open software development to non-engineers and redefine who gets to build in the AI era, challenging Indian IT’s agent-led status quo with a more inclusive, creative future. While Big Tech views India as a market, Perplexity sees it as a mission, betting on Bharat with free student access, Airtel and Paytm tie-ups, and a shift from chatbots to agentic search, all while staying fiercely independent in a world ruled by tech giants. LinkedIn gave the world Kafka; now it’s outgrowing its own creation, replacing it with Northguard and Xinfra—a self-healing, cloud-native pub-sub layer built for 32 trillion daily events and a future Kafka was never designed to handle. Now, let’s explore some exciting collaborations and exclusive insights from the AIM ecosystem, brought to you with a unique twist outside our standard editorial content. In a world where your job could change before your title does, Tredence is building a talent engine that learns faster than technology evolves, blending bootcamps, live experiments, and role transitions into a system built for the AI era. Learn more here. What if your voice could be understood, without changing who you are? Sanas is tackling accent bias with real-time Speech AI that preserves identity while improving clarity, and India might just be its biggest proving ground. Read the full story. AI agents aren’t just automating tasks—they are reasoning, learning and collaborating with humans to reshape how work gets done. This deep dive by ValueLabs explores how agentic AI is redefining enterprise roles, and why the future of work is ‘Human + AI’, not human vs AI. Read more here. As GBS and GCCs outgrow back-office roles, this report reveals why platform-led, AI-first models, powered by Agentic AI, are redefining enterprise value creation, from incremental automation to intelligent reinvention. Read the full EdgeVerve-SSON collaborative report here.  Every employee managing a team of AI agents isn’t sci-fi anymore; it’s the foundation of AI-native thinking. At MachineCon 2025, UnifyApps and Virtusa unpacked how India’s GCCs are leading this shift from tool adoption to enterprise reinvention. Read the full story.   Now that you are caught up on the top headlines and insights from our partner ecosystem, let’s shift gears, because one question continues to dominate boardrooms: what truly makes a workplace great for data and AI professionals? AIM’s newly released list of Best Firms answers just that, spotlighting organisations like TheMathCompany, CEAT, AB InBev and Wipro, each recognised for building purpose-driven, people-first AI workplaces, backed by employee sentiment and independent benchmarks. Learn more here. [Watch Now] Moreover, speaking of building for the future, Google just dropped its biggest India-first AI announcements at I/O Connect 2025 in Namma Bengaluru, and AIM Tv was at the centre of it all. From real-time demos with Gemini 2.5 to exclusive conversations with Google DeepMind and the team behind Gemma, we captured what Google is really building for India’s developers.