The question I get asked most often at events isn’t about placements or fees. It’s a version of: why a new B.Tech? What was wrong with what already existed?
The honest answer starts before Kalvium.
Where the question came from
Rajesh and I previously co-founded FACE Prep. Fifteen years inside engineering placement-prep. We worked with more than 2,000 colleges across India and reached 60 lakh+ students preparing for jobs at the end of their B.Tech.
That seat gave us a view most edtech operators don’t get. We weren’t selling content. We weren’t running marketing campaigns at the top of the funnel. We were sitting at the very end of the four-year B.Tech, three to six months before graduation, looking at what the system had produced.
What we saw, year after year, was the same thing.
Students arriving with grades. Students arriving with certificates. Students arriving without the ability to ship a working piece of software end-to-end. The students weren’t the problem. They’d done what was asked of them. The structure they’d just spent four years inside hadn’t asked the right things.
By the time a candidate reached us at the end of college, the gap was real and we could measure it. The job-readiness training we ran was a patch. A useful patch, but a patch. It didn’t fix what the four years had failed to build.
After enough of that, you stop wanting to patch.
The decision
Around 2021, a few of us started talking seriously about a different version of the question.
What if the four years themselves were different? What if production code started in year two, not year four? What if the programme were designed by people who’d seen what hiring managers actually look for, sitting inside the structure that produced the candidates?
You can’t fix engineering education from the outside. We tried for fifteen years from the outside. So we decided to build it from the inside.
That’s what Kalvium is. It isn’t a placement-prep company. It isn’t a coding bootcamp. It’s a B.Tech CSE programme designed by people who’d watched the existing four years fail to build what employers need, and who decided to use the apprenticeship that’s been in India’s books since 1961 to do something about it.
Why the four of us
Kalvium has four co-founders. Each brings a different angle on the same problem.
Rajesh and I bring the 15 years of seeing the gap from the placement-prep side. We know what the end of college looks like. We know what makes the difference between a hireable engineer and a graduate. That’s the deepest version of “what should the four years build for.”
Anil came from Google and HackerRank. He’s seen what production engineering teams actually need, both inside a global product company and across the assessment infrastructure that hires for one. He brings the hiring-side rigour.
Deepak’s background is in systems thinking and large-scale operations. He’s the one who keeps asking: does this scale? Will it work at 200 students per campus, then 500, then 2,000? Rigour at one squad means nothing if it falls apart at fifty.
Four angles, one problem. Most edtech founder teams are missing at least one of them.
Why partner universities and not our own campus
This is the second question I get at every event, usually right after the first.
The honest answer is operational. Building a new university in India is a ten-year regulatory project. Land, accreditation, UGC, AICTE, state government, the whole stack. Building a programme that runs inside nine existing universities is a four-year operational project. We chose the operational route because students who’d be 22 by the time a new campus opened deserved the programme now, not in a decade.
The split works because the responsibilities split naturally.
The partner university handles what universities are built to handle. Accreditation, the physical campus, hostel, the degree-issuance side of higher education. Nine of them: AMET in Chennai, JECRC in Jaipur, Kalasalingam in Krishnankoil, LPU in Punjab, SGT in Gurgaon, SRM University AP, St Joseph in Chennai, Yenepoya in Mangalore, and Yenepoya in Bangalore. Plus the Kalvium Direct Online variant.
Kalvium handles the programme. Pedagogy, mentorship, the squad structure, the DOJO coding belts, the partner-company integration, the placement support, the assessment design. Fee is paid to the partner university directly. Kalvium does not collect tuition. The only payment that comes to Kalvium is the ₹10,000 seat reservation fee at admission.
People sometimes read this as a workaround. It isn’t. It’s the design.
Why production code starts in year two
This is the one that hiring managers understand first and parents take a little longer to.
The forgetting curve, mapped 140 years ago by Hermann Ebbinghaus, doesn’t lie. Students forget most of what they learned in year one by the time year three arrives, unless something specific is done about it. The fix isn’t more lectures. It’s earlier application. Material that gets used stays. Material that doesn’t get used doesn’t.
The conventional B.Tech runs the application at the end. The capstone project sits in year four. By the time a student gets there, the foundational courses they need to apply are eighteen months gone. So most capstones become eight weeks of relearning, then four weeks of building something small.
At Kalvium, capstone projects start in semester two. Production work with partner companies starts in year two. Thirty to forty hours a week, on real codebases, for thirty-six months. By the time a Kalvium student finishes year four, they’ve shipped more than 4,000 tracked hours of real work with real teams.
That changes what year four looks like. It changes what the resume looks like. It changes what the hiring conversation looks like, because the company isn’t being asked to bet on a transcript. They’re being asked to look at months of verified output.
We didn’t invent any of this. The apprenticeship principle has been in the books for 63 years. What was missing wasn’t the policy. It was the operating layer that could actually track and coordinate work-integrated learning at scale.
We built that layer. The rest is the consequence.
What we don’t promise
A short list, because it matters.
We don’t promise placement. No programme can. We do publish what the cohort has done. Batch 2026: 82.4% placed as of March 2026. The batch hasn’t graduated yet, so the number is still moving. Median package ₹16.5 lakh per year. Lowest ₹15 lakh. Highest ₹36 lakh. Named hiring partners: Morgan Stanley, PhonePe, Thoughtworks, Lowe’s, Tata 1mg, Maersk, Yellow.ai, and forty more.
We don’t promise that every student will succeed at the level of Vidvath J., the first-year in Squad 49 who got selected for Google Summer of Code out of 43,984 applicants from 172 countries. Or that every team will be Squad 56, who won the Smart India Hackathon 2025 Grand Finale. These outcomes happen because the structure made them possible, not because Kalvium handed them out. The students did the work.
We don’t promise that Kalvium is for every student. It works well for the ones who can stay with a difficult problem, learn quickly, and treat building as the medium of learning. It works less well for students looking for a chill four years. We say that out loud at every event.
What we do say: this is a hard programme. The bar to get in (KNET) is fit-based. The bar to get through it is real. The students who graduate finish as engineers, not as graduates with a CGPA.
Three and a half years in
We’re three and a half years into the bet. There are 2,223 students currently in the programme, across nine campuses. The first full graduating cohort is six months away. The Career Advisory Board includes Nithin Kamath at Zerodha, Kunal Shah at CRED, Rahul Chari at PhonePe, Ankit Bhati at Ola, Sarv Saravanan at Microsoft, and Abhilash Nair at Google. None of them are paid to be there. They’re there because the version of B.Tech we’re building is one they’d want their own kids to do.
The hardest part of the next three years is operational. Holding the rigour at scale. Making sure Squad 73 in 2028 is the same programme Squad 49 is in 2026. Not letting the partner network drift. Keeping the apprenticeship promise honest, week after week, across nine campuses.
I think we’ll do it. I also think we’ll make mistakes we haven’t seen yet. The version of Kalvium that exists in 2030 will be different from the version that exists today. That’s the point. A programme that doesn’t change isn’t paying attention.
What I tell families at the booth is the simplest version of all of this. Engineering education in India has a structural problem. We’ve spent fifteen years watching it from one side and three and a half years rebuilding it from the other side. We don’t have it all figured out. We do have a programme that is, by the evidence of the cohorts that have gone through it, doing the thing it claims to do.
If your child is in the room asking the kind of question that makes you proud, take the KNET. Come visit a campus. Talk to a Squad-49 student about what their week looks like. Ask the questions that worry you.
We’ll be honest about every one.
Onwards.