There are two stories running across India's tech industry right now, and on the surface they contradict each other.
The first story is about layoffs. They are real, they are happening across IT services, and in more than a few cases AI is being named as the reason.
The second story is about scarcity. The same industry that is letting people go is also saying that India's AI buildout — the thing everyone agrees is the future — is being held back by a lack of skilled talent.
So which is it? Too many engineers, or too few?
I talk to engineers every single day, and from where I sit the answer is neither. We are not short on supply. We are short on the right alignment between the talent we already have and the work that is actually being created.
Where the work has actually gone
The clearest signal is in who is hiring.
In FY26, Global Capability Centers (GCCs) added roughly 200,000 jobs in India, while IT services firms added around 110,000, according to data from staffing specialist Xpheno reported by The Economic Times. This is the third year running that GCCs have out-hired the services sector, and the gap is widening — not narrowing.
That isn't a rounding error. It's a structural shift. The center of gravity of Indian tech employment is moving from the services model to in-house capability centers — and those centers are building, not just maintaining. A large share of their new roles are tied directly to AI, data, cloud, and platform engineering.
So the demand is enormous. The pain engineers feel is real. But the two are not opposites. The work has moved. Most engineers simply haven't moved with it yet.
That is what I mean by an alignment problem. And alignment, unlike a genuine talent shortage, is fixable — and far faster than most people assume.
The shortage is real on paper — and that's the opportunity
None of this is to wave away the skills gap. By most measures it is severe. Industry surveys through 2025 and into 2026 put India's AI skills deficit somewhere around half of demand, with even sharper shortfalls in generative-AI-ready talent and in the mid-to-senior cohort where deep technical skill has to combine with delivery ownership.
But here's the reframe: a 50% "deficit" against exploding demand is not a verdict that India lacks capable engineers. India has one of the largest, sharpest engineering workforces on the planet. The deficit exists because the workforce is calibrated to yesterday's roles while the demand curve has jumped to AI-native ones.
Close that calibration gap and the same number that reads as a crisis today starts to read as the largest skilling opportunity Indian engineering has ever had.
Two shifts that get us there
Moving an engineer from "at risk" to "in demand" takes two shifts, and the order matters.
The psychological shift comes first. Many engineers are still waiting for the old roles to come back, treating the disruption as something to survive rather than something to ride. The mindset has to flip: the GCC AI buildout is not a threat to your career, it is the single biggest opportunity in front of it. The people who internalize that early will be the ones who get there first.
The technical shift comes second. Once the mindset is right, the work is to align skills to what these centers are actually building — not to a generic notion of "AI," but to the specific AI, data, and platform capabilities GCCs are hiring for right now.
"Just learn AI" is not guidance
Here is where the system is failing engineers today.
The advice on offer is almost always some version of "learn AI." That is not guidance. That is a shrug. It tells an engineer nothing about what to learn, in what order, or — most importantly — how to get real, hands-on exposure rather than another certificate that proves nothing.
There are very few practical, targeted courses built around what GCCs are genuinely hiring for. And there is very little honest awareness-building. So motivated, capable engineers are left to guess, and guessing is slow and demoralizing.
There's a deeper version of this gap inside companies, too. For most teams, "using AI" still means bolting it onto code generation, code review, and a handful of automations. That's useful, but it's narrow. Almost nobody has built a framework for how every engineer works with AI across the full development lifecycle. Until that happens, both individuals and organizations are leaving most of the value on the table.
From shortage to surplus in under a quarter
Put the pieces together and the conclusion is more optimistic than the headlines suggest.
If we pair the psychological shift with targeted, practical skilling — small courses built around real GCC AI requirements, with genuine exposure rather than theory — I'm convinced we can move India from "short on supply" to a surplus of the right-skilled talent in less than a quarter.
And the most powerful part: we don't get there by producing more engineers. We get there by aligning the ones we already have.
That conviction is the entire reason I'm building MyJobFriend — to give job-seeking engineers the specific, practical guidance the market keeps promising and failing to deliver: what to learn, in what order, and how to build the exposure GCCs are actually looking for.
If you're an engineer feeling the squeeze right now, don't read the layoffs as the end of the story. Read them as a signal to realign. The demand is waiting. The question is whether your skills are pointed at it yet.
Want help aligning your skills to where the AI jobs actually are? Explore myjobfriend.in
Sources: GCC vs IT services FY26 hiring — Xpheno data via The Economic Times / Business Standard. AI skills-deficit figures — industry surveys (NASSCOM–Deloitte, EY GCC Pulse) reported through 2025–2026.