Fresher Hiring Just Split in Two — Most Screening Rubrics Weren’t Built for Either Half

Fresher hiring (0–3 years) grew 16% year-on-year in early 2026, faster than any other experience band in the Indian job market, according to Naukri’s JobSpeak report for January 2026. That headline sounds like good news for anyone running a campus program. It isn’t the whole story, and the part it leaves out is the part that breaks most screening processes.

Two very different jobs are hiding under one label

By March 2026, the same Naukri JobSpeak data showed demand for freshers in the 20+ LPA salary band up 23% year-on-year, with AI/ML specialist roles seeing the sharpest pull toward the 50+ LPA band. In the same reporting window, large IT-services majors restarted mass campus hiring at starting packages closer to the well-documented ₹3.5–8 LPA entry-level IT band, with mass-hire offers clustering at the low end of that range. Both of these candidates carry the word “fresher” on their offer letter. Neither has done a single day of paid work. That’s where the similarity ends.

A ₹3.5 LPA mass-hire drive is a volume problem: tens of thousands of applicants, a narrow band of required skill, and a screening job that’s mostly about triage — can this person communicate clearly, show up on time, and hold a structured conversation. A 20+ LPA specialist-fresher drive is closer to a technical hiring problem wearing a campus badge: the applicant pool is smaller, but the bar for domain depth is much higher, and a generic “can you communicate” screen tells you almost nothing about whether someone can reason through a GenAI systems question.

Most Indian TA teams built one fresher screening rubric years ago and never split it. That was fine when “fresher” meant roughly the same job at roughly the same pay everywhere. It stops being fine the moment the same TA org runs a 3,000-candidate mass campus drive and a 40-candidate specialist-AI campus drive in the same recruiting season, using the same intake form and the same interview questions for both.

Why the mismatch is invisible until it isn’t

The failure mode doesn’t show up as an obvious error. It shows up as noise. Run the mass-hire rubric on specialist candidates and you’ll pass people who communicate well but can’t do the work, because the rubric was never built to probe technical depth. Run the specialist rubric on the mass-hire pool and you’ll reject good communicators for not name-dropping frameworks they were never expected to know, because the bar was calibrated for a different population entirely. Both mistakes look like “our screening is a bit noisy” from the outside. Neither gets fixed by tightening the same rubric further — tightening a rubric built for the wrong population just makes the wrong population pass more consistently.

This is the piece the “fresher hiring is booming” headline skips. Volume growth is a capacity question, and Indian HR teams have gotten reasonably good at solving capacity — we’ve written about the arithmetic of running a 3,000-candidate drive in one evening. Composition growth, two genuinely different candidate populations sharing a job title, is a taxonomy question, and taxonomy problems don’t get solved by running the same process faster.

What actually changes at the screening layer

In HireQwik’s pilot data across 1,099 interviews spanning 14 campaigns, the common thread wasn’t a single rubric — it was communication-first screening applied to the right bar for the role. A structured 15–20 minute voice conversation that filters for communication and reasoning, not keyword-matched resumes, works for the mass-hire tier because it catches the candidates who look identical on paper but can’t hold a coherent conversation. It only works for the specialist tier if the bar for that conversation actually reflects specialist-level reasoning, not the same set of generic questions calibrated for entry-level volume roles.

None of this requires guessing which tier a candidate belongs to. The JD already tells you: a ₹3.5 LPA mass-hire req and a 50+ LPA AI/ML specialist req are visibly different documents before a single resume arrives. What breaks is when the screening process behind both reqs quietly reuses the same question bank and the same pass bar because nobody revisited it after the fresher market split.

The contrarian take: treating “fresher hiring” as one problem in 2026 solves neither half of it well. If your screening process can’t tell a ₹3.5 LPA campus hire from a 50+ LPA specialist fresher before the interview even starts, you’re not running one rubric efficiently — you’re running the wrong rubric on half your candidates, and calling it consistency.

The honest limit

We don’t claim an AI screen can look at a fresher’s resume and correctly infer which tier a role belongs to on its own — that’s a call the JD and the hiring manager make, not the screening layer. What the screening layer can do is stop treating “0–3 years experience” as a single population once the role, and the pay, say otherwise.

If your campus season is running both tracks at once this year, see how a communication-first screen prices out for your specific applicant volume.