From Saltmarch
Week of 13th April 2026

India’s software industry is not slowing down in any obvious way, but it is no longer behaving in a manner that can be read through the signals that worked for the last two decades.

For a long time, the logic was simple. Headcount growth tracked demand, and demand translated into revenue. That relationship is now under strain.

Tata Consultancy Services ended FY26 with a decline in annual revenue of ~0.5% in dollar terms while simultaneously reporting annualised AI revenue of over US$2.3 billion in Q4 [1]. That combination, more than any headline, captures the shift underway.

The broader sector is moving in the same direction, even if the specifics vary.

Infosys has continued to grow through large deals and selective hiring rather than broad-based expansion [2]. Wipro has seen muted revenue growth but improving margins, reflecting tighter cost discipline [3]. HCLTech has begun to explicitly tie growth to AI-led services, reflecting a shift in where future demand is expected to come from [4].

This is not contraction. It is compression. The system is being asked to deliver more without expanding in the same way.

Where the Pressure Is Coming From

What makes this moment different is that the pressure is not coming from a single source. Part of it is technological. AI-assisted development is beginning to reduce the effort required across coding, testing, and migration. The gains are uneven, but at scale they are material. Part of it is commercial. As one industry analyst noted, clients are no longer buying effort; they are increasingly demanding outcomes [5]. And part of it is structural. Modern engineering environments make productivity visible in ways that were not possible earlier. That visibility changes how organisations think about staffing.

When these forces combine, headcount stops being the primary lever for growth.

AI in Production, and the Questions It Raises

There is very little ambiguity about adoption. AI is now firmly in use across enterprises, and importantly, it is moving from experimentation into scaled deployment.

Tata Consultancy Services itself described FY26 as a “pivotal year for enterprise AI adoption”, with deployment accelerating across industries [6].

But adoption is not the same as value.

A second-order issue is emerging, and it is already visible in engineering teams. Cost.

• workflows that scale unpredictably
• tools that are invoked more often than expected
• context layers that increase inference cost
• agent-based systems that add complexity faster than they add value

The constraint is no longer just capability. It is control. The teams that are navigating this well are treating AI systems as infrastructure, not features.

The GCC Opportunity Is Changing Shape

India’s Global Capability Centre ecosystem continues to expand, and remains one of the strongest structural drivers of engineering demand. But the nature of work within these centres is shifting.

A growing share of existing work is susceptible to automation through AI, particularly in areas that are repetitive or rules-driven. At the same time, organisations are moving higher-value work into India, including platform ownership, architecture, and data systems.

The direction is clear. There is less emphasis on execution that can be standardised, and more emphasis on systems that require judgement, ownership, and long-term thinking. The opportunity is expanding, but not evenly.

Hiring: Strong on Paper, Selective in Practice

On the surface, hiring intent remains positive. Companies continue to recruit, and large deals are still being signed. But the underlying trend is more nuanced.

Net hiring across the top Indian IT firms has slowed dramatically, with almost no net additions across the largest firms over much of FY26, compared to strong hiring in the previous year [8].

This does not mean demand has disappeared. It means it has narrowed.

Roles that continue to see demand:

• cloud and platform engineering
• data systems
• AI and ML with production exposure
• security and reliability

Roles that are weakening:

• generic development roles
• support-heavy delivery
• low-complexity execution work

The gap between available jobs and suitable candidates is widening.

Capital Is Being Allocated Differently

The funding environment tells a similar story of selectivity. While total funding has moderated, early-stage investment remains active, and large deals are still being signed across the IT services sector.

For instance, Infosys alone reported large deal bookings of US$4.8 billion in a single quarter [9].

What has changed is not the availability of capital, but the conditions attached to it. Investors are still willing to fund new ideas. They are far more cautious about funding scale without proof.

A Number That Matters

2.3 billion

That is the annualised AI revenue reported by Tata Consultancy Services in Q4 FY26. It is still small relative to total revenue, but it is large enough to indicate that AI has moved beyond experimentation into meaningful commercial activity [10].

What Is Changing, Quietly

The most important changes in Indian software right now are not dramatic, but they are visible in how decisions are being made. Firms are hiring more selectively. Clients are demanding clearer outcomes. AI is beginning to affect both delivery and pricing, and Investors are placing a higher premium on proof.

None of this points to a slowdown. It points to a system that is becoming more disciplined.

The next phase will not be defined by how quickly software can be built, but by how effectively it can be delivered, operated, and justified.

From Saltmarch

This article is part of a series tracking the Indian developer ecosystem.

Each edition will focus on what is actually shifting beneath the surface:

• how large firms are adapting
• how AI is behaving in real systems
• where hiring is moving
• how capital is being deployed