Engineering placement season used to be a volume exercise. A company arrived, ran an aptitude test, made several hundred offers at a single salary, and worked out who could actually do what somewhere around month four of training.
That model is coming apart, and AI is the reason.
Market data indicates that differential salary for specialised skills now accounts for roughly 30 to 40 percent of overall campus hiring, and is expected to reach 45 to 55 percent in 2026-27. Read that again slowly. Within about a year, more than half of campus offers may be priced individually rather than by batch.
IT services firms have pulled back on mass recruitment from engineering colleges while simultaneously increasing hiring of graduates who show up with AI, cybersecurity, cloud or data engineering skills, at noticeably higher salaries. Same campuses. Very different offer letters.
Volume did not vanish. It reorganised.
TCS still plans to onboard around 42,000 entry-level hires in FY26. Infosys is targeting more than 20,000 freshers, Wipro somewhere between 10,000 and 12,000 across services, consulting and newer technology roles. Around 73% of Indian employers indicated plans to hire freshers in the first half of 2026. This is not a freeze.
What changed is the sorting. A drive that once produced one offer letter now produces three tiers, and the distance between the top and the bottom tier is wide enough to shape a career from the first year onward.
The India Skills Report 2026 put overall employability at 56.35%, a steady climb over the past decade and still a number meaning close to half of graduating candidates are not considered work-ready. When employers start paying a visible premium for demonstrable AI capability, that half does not stay level. It falls further back.
Portfolios beat certificates now
Employers screening freshers increasingly ask for projects, portfolios and evidence of ability rather than course completion certificates. Reasonable, given that every student in the country can now finish an AI course. Very few can explain a design decision they made under a real constraint, or describe something that did not work.
This is a better filter than an aptitude test. It is also a more unequal one. Building a substantial project takes compute, unstructured time, mentorship and usually some amount of family cushion. Tier-1 campuses supply that environment almost by accident. Most colleges do not supply it at all.
Companies hiring in smaller volumes have started routing early-career sourcing through external AI recruitment services instead of sending their own teams to fifty campuses. Efficient, certainly. It also narrows the funnel to whichever institutions those firms already cover, which tends to be the same fifteen colleges everyone else is fighting over. Worth knowing what you are trading away when you buy the convenience.
The build side of the equation
Buying AI-skilled freshers is only half of what the large employers are doing. TCS trained around 350,000 employees and Wipro around 220,000 on AI during 2023-24. Work by Deloitte and Nasscom notes Indian firms have begun internal upskilling at serious scale, and that roughly 60% of workers, rising to 71% among Gen Z professionals, believe AI skills improve their career prospects.
That appetite is an asset most countries do not have. The open question is whether those programmes are producing capability or producing completion rates. It is the same problem the campus filter is trying to solve, just relocated inside the building where it is harder to see.
Three things worth doing if you hire freshers
Decide what a fresher AI role actually involves before you price it. A lot of roles labelled AI consist of data cleaning and prompt iteration. Honest work, useful work, but paying an AI premium for it and then losing the person at month eight when they figure it out is an expensive way to learn.
Widen the search. Tier-2 and tier-3 institutions have students with genuine projects and no placement pipeline reaching them. Screening costs more. Retention is usually better, and the compensation expectations are not inflated by four competing offers.
Write the job description honestly. The distance between what an AI role was sold as and what it turned out to be is doing more damage to early-career retention in Indian tech right now than compensation ever did.
Author Bio
Nikhil Vaidya
Nikhil Vaidya is the CEO of Prism HRC, a leading recruitment services company in India. Nikhil's expertise in talent acquisition and has been instrumental in connecting hundreds of top-notch clients with exceptional IT talent over the last 15 years.
Business Outstanders brings you sharp insights on tech, business, entrepreneurship, law, crypto, and more. We uncover what’s next. Stay updated, sign up for our newsletter and be part of the future!
