I’ve been on the hiring side for a while. At HackerRank, I built and led engineering teams. At Google, I ran hiring loops. Now I’m at Kalvium, working on what happens earlier: the four years that determine what kind of engineers those loops are going to see.
One pattern stays consistent across all of that. The colleges that produce engineers companies actually want to hire aren’t the ones with the most impressive-sounding syllabus, or the longest career advisory board list. They’re the ones whose curriculum was shaped by people who are still making hiring decisions.
Most B.Tech programmes have advisory boards. Most are ceremonial. A few marquee names on a brochure, a guest lecture once a year, maybe a half-day workshop. The syllabus moves on a review cycle that’s three to five years behind what the industry evaluates right now. That’s not negligence. It’s structural. The people designing the curriculum aren’t the people running interviews this quarter, and they haven’t been for years.
The result is the gap I keep seeing. Candidates who know the material. Candidates who can clear the technical screen. But candidates who’ve never worked on a real codebase, never owned something that could break on real users, and never had to figure out what to do next when nobody was pointing the direction. Four years of assignments graded by a teacher and then archived. The accountability muscle never builds.
What Kalvium’s board actually is
It isn’t ceremonial.
Kalvium’s Career Advisory Board has 24 members who are currently running engineering teams, building products, or making hiring decisions. Their input runs through the actual curriculum design: the DOJO problem sets, the project briefs students work on, the two specialisation tracks, and the Mastery layer that defines how industry-relevant skills get evaluated.
Here’s who that includes.
Nithin Kamath, CEO of Zerodha. Kunal Shah, founder of CRED. Rahul Chari, co-founder and CTO of PhonePe. Ankit Bhati, co-founder of OLA. Aravind Sanka, co-founder and CTO of Rapido. Sarv Saravanan, VP at Microsoft. Abhilash Nair, APAC Recruiting Leader at Google. Sujeet Kumar, co-founder of Udaan. Aakrit Vaish, CEO of Haptik. Amar Prabhu, co-founder and CTO of Rupeek. Anupam Mittal, founder of People Group. Mathangi Sri, Chief Data Officer at YUBI. Gazal S. Kalra, Head of Strategy and Operations at Yellow.ai. Neel Kothari, CTO at Bajaj Finserv Health. Kumar Muthuraman, Professor at the University of Texas at Austin. Guru Venkatachalam, VP and CTO at VMware. Kulin Shah, co-founder of Onsurity. Anurag Vohra, CTO at TMRW. Muralidharan, Advisor at Switch Labs and formerly of Rivigo. Manjunath Talwar, CEO of iZoot. And Anil Goel, President of Technology at BYJU’s.
That’s not a brochure list. These are people hiring CSE engineers this year, or shaping what their companies need in the next three years.
What the board’s input actually produced
Let me be specific.
The Mastery layer sits outside written exams. In most B.Tech programmes, degree assessment and skills assessment are the same thing. Pass the exam, you’re certified as knowing the subject. The problem is that companies don’t evaluate engineers through written exams. They use code review, system design conversations, production deployments, and judgment under real constraints.
Kalvium’s curriculum has three layers. Essentials is the AICTE-compliant degree structure, assessed through university exams as required. Mastery is industry-aligned capability, verified through codebases, deployments, and system artifacts, not written tests, because that’s how the board members’ companies actually evaluate engineers. Excellence is ownership: open-source contributions, mentoring, and real products built and shipped.
The Mastery layer is kept outside degree assessment so it can evolve with industry standards. When what the market evaluates changes, the Mastery layer changes. A written-exam assessment doesn’t work that way.
The two specialisation tracks reflect where hiring is headed. The board shaped what the market would need at graduation time, not what the programme found easiest to teach. That produced two tracks: Software Product Engineering for students going deep on full-stack, system design, and applied AI, and AI and Future Technologies for students focused on machine learning systems, mathematical foundations for ML, and AI engineering.
Both tracks share the same first two semesters and the same Essentials layer. The divergence from Semester 3 onward is based on where hiring demand is now and where it’s going. The track decision post covers the specific subject differences in detail if you’re deciding between them.
The DOJO is calibrated to real interview problems. Daily practice using problems from real coding interviews and competitive platforms. Six belts per language: Java, C++, JavaScript, Python. Measurable progress from Semester 1.
The difference between a student who’s been working through real hiring-test problems six days a week for two years and one who’s been doing textbook exercises is real and shows up immediately in a hiring loop. That calibration came from board members who sit on both sides of that table.
What the companies who hired the graduates said
Here’s where the co-design shows up in output, not just input.
Sarv Saravanan at Microsoft specifically named development style and environment, not programming language syntax, as what Kalvium’s programme gets right. Companies don’t hire for syntax knowledge in isolation. They hire for someone who can work inside a real codebase, with other people, under a deadline, when the right answer isn’t obvious.
Amar Prabhu, co-founder and CTO at Rupeek, described Kalvium students as combining technical knowledge with hands-on experience and the ability to handle challenges with confidence and adaptability. That’s not a generic description. It’s what you get from two or more years of production-adjacent work reviewed by someone who treats you as a colleague, not a student.
Aswath Premaradj, founder of Zentience, put it plainly: the students’ full-stack engineering skills were strong enough that selection was genuinely difficult. He could only take four.
Three companies, different types and sizes, consistent descriptions. That’s a pattern worth examining.
The actual measure
Whether curriculum co-design worked is in the outcomes.
As of March 2026, 82.40% of Batch 2026 was placed, with the batch still months from graduation. The median package was ₹16.5 LPA. The floor was ₹15 LPA. The highest was ₹36 LPA. Recruitment came from Morgan Stanley, PhonePe, Thoughtworks, Lowe’s, Yellow.ai, Tata 1mg, Maersk, Medable, Rupeek, and 40 more. 28% of placed students took international roles. The programme runs across nine partner universities for Admission Year 2026-27, spanning Karnataka, Tamil Nadu, Andhra Pradesh, the NCR, Punjab, and Rajasthan.
The advisory board is part of why that outcome exists. Not because names on a document change what students learn, but because people who are currently hiring explained what they evaluate, and the programme built that into the daily practice from Semester 1 onward.
For the year-by-year build arc of what that curriculum produces in practice, What a Kalvium student builds in four years walks through the arc from Year 1 to Year 4.
For the hiring-side view on why production experience matters more than the syllabus, The hardest part of building engineers, from the hiring side covers the gap most programmes never close.
For placement numbers with full context on how to read them honestly, Kalvium placements: how to read our numbers honestly is the right place to start.
Anil is a co-founder of Kalvium and previously led engineering teams at Google and HackerRank. He writes about what the Indian tech market actually rewards, from the interview side of the table. Read more from Anil or browse the btech category.