The question I hear most from families at orientation is some version of this: what exactly does a student build over four years?
It’s a fair question. Most engineering programme materials describe aspirations, not specifics. They mention real-world projects, industry-integrated learning, hands-on curriculum. None of that tells a family what their child will actually be building by Semester 2’s end, or what changes between Year 1 and Year 3.
I lead programme design and delivery at Kalvium. What follows is the honest answer, grounded in the curriculum structure, not the prospectus.
Why a build arc works differently from a subject list
The distinction matters.
In 1991, Phyllis Blumenfeld and her colleagues at the University of Michigan published a paper in Educational Psychologist that drew a line worth keeping in mind here. They distinguished between “doing activities” and genuine project-based learning. Their finding: projects produce real learning when they involve a meaningful driving question, sustained inquiry over time, and an artifact that someone outside the classroom can examine and critique. Doing an activity doesn’t guarantee any of those things.
Why does this matter for a walkthrough? Because what a student builds isn’t just the output of the subjects they’ve taken. It’s the result of how the complexity scales across years and how the work shifts from structured exercises to open-ended problems. The arc is deliberate, not incidental.
Year 1: The web stack, from the start
Both semesters here carry 24 credits each.
Semester 1 covers seven subjects: Front-end Web Development, Problem Solving using Programming, Critical Thinking, English LSRW, Design for Developers, Breadth of CS-1, and Learning How to Learn.
That last one matters. The programme treats the meta-skill of learning how to acquire hard skills as a subject in its own right. It’s in Semester 1 because it’s needed from Semester 1, not because it was convenient to place there.
Front-end Web Development isn’t deferred to Year 2 or 3. Week 1. Students write and ship code from the first week, which puts deliberate practice at the beginning of four years rather than at the end of them.
DOJO also starts in Semester 1. Six days a week, daily coding practice with a belt system across Java, C++, JavaScript, and Python, six belt levels per language, each earned through testing rather than attendance. The daily practice structure doesn’t change across four years. What the student brings to each session does.
Semester 2 builds the other half of the web stack: Back-end Web Development, Practical Databases for Web Development, Introduction to AI/ML/LLMs, Discrete Mathematics, Professional Skills for the Workplace, Breadth of CS-2, and Indian Constitution. Another 24 credits, same across both tracks.
By the end of Year 1, students have built full-stack web applications. A front end, a back end, and a real database connected together. For the full argument on why this starts in Year 1 rather than later, full-stack development in the first year makes the case in detail.
Year 2: First real projects and the beginning of the track split
Year 2 is where the two specialisation tracks diverge for the first time.
Semester 3 carries 22 credits. Both tracks study Database Management Systems, Object-Oriented Programming, Introduction to Generative AI Systems, Values and Ethics in Technology, Environmental Sciences, and the first Internship or Project module. The first split: Software Product Engineering students take Financial Literacy; AI and Future Technologies students take Mathematical Foundations for ML instead.
The Internship or Project 1 module is the structural change Semester 3 introduces. Before this point, every piece of work had defined parameters and a correct output. From Semester 3, the brief is the starting point, not the boundary. Students move into one of four work tracks: Simulated Work, Internship, Open Source Projects, or Build AI-native products. What Semester 3 at Kalvium actually involves covers this shift in more detail.
Semester 4 carries 21 credits: Data Structures and Algorithms, Operating Systems, Deep Learning and Neural Networks, Project Management and Agile Delivery, Organizational Behavior, and Internship or Project 2. Both tracks cover the same subjects.
The sequencing of DSA in Semester 4 isn’t arbitrary. Students arrive at it having already written real code daily for well over a year and having built front-end and back-end systems connected to real databases. When DSA arrives, it arrives in a context where the student has already encountered the problems it solves. That’s a different entry point than studying it in isolation at the start of the programme.
Year 3: Specialisation deepens and systems-level thinking arrives
By Year 3, the two tracks are running on clearly different rails.
Semester 5 carries 21 credits. Both tracks study Design and Analysis of Algorithms, Computer Organization and Architecture, Universal Human Values, Indian Knowledge Traditions, and Internship or Project 3. The split: Software Product Engineering students study Applied AI and Agentic Systems; AI and Future Technologies students study Machine Learning Algorithms.
Semester 6 carries 20 credits. This is where the tracks diverge at the core technical subject. Software Product Engineering students study System Design. AI and Future Technologies students study AI Systems Engineering. Both tracks also cover Computer Networks, a Professional Elective, Career Development and Professional Readiness, and Internship or Project 4.
System Design in Semester 6 is where five semesters of prior building start to cohere into something a student can reason about architecturally. By Semester 6, a student has designed database schemas, built backend systems, worked through four project modules, and studied algorithms and operating systems. They’re in a different position to engage with system design because of it. The academic weight of the subject is the same. The starting point isn’t.
Career Development and Professional Readiness in Semester 6 isn’t padding. By this point in the programme, some students are a few months from a placement conversation. The technical skills from the previous five semesters are necessary for that. This subject builds the professional communication and readiness layer that makes them sufficient.
Year 4: Electives, entrepreneurship, and the Final Capstone
Semester 7 carries 16 credits: Professional Electives 2 and 3, Business Thinking and Entrepreneurship, and Internship or Project 5. The AI and Future Technologies track takes AI-focused professional electives; the Software Product Engineering track has the broader elective set.
Semester 8 carries another 16 credits: an Open Elective and the Final Internship or Capstone Project. AI and Future Technologies students take an AI/Tech-focused Open Elective.
The Final Capstone is what Blumenfeld and her colleagues described as the external artifact that transforms project work from an exercise into a genuine learning event. It’s evaluated by someone who didn’t design the brief: a technical panel, a prospective employer, a mentor assessing whether the work holds up to scrutiny. That’s a different kind of accountability than a course exercise where the assessor already knows what the correct answer looks like.
The Capstone is also the culmination of a build arc that started in Semester 3, not a one-semester sprint at the end of Year 4. By Semester 8, a student has already completed five prior Internship or Project modules, each building on what the previous one produced. For the design logic behind starting capstone-style work early rather than concentrating it at the end, why capstone projects start in Semester 2, not Year 4 explains it directly.
The three layers that run alongside the year-by-year picture
The curriculum guide describes three layers running in parallel across all four years.
Essentials covers the formal AICTE-compliant degree subjects, assessed through university examinations. Total: 164 credits. These are what the degree itself represents.
Mastery covers employability. Industry-aligned work verified through codebases, deployments, and system artifacts rather than written exams. It’s kept outside the formal degree structure deliberately, so it can update with industry expectations rather than wait for the next curriculum revision cycle. The curriculum guide’s framing is direct: “Essentials show what a student has studied. Mastery shows what a student is ready to do.”
Excellence covers ownership: open-source contributions, mentoring, deep technical writing, and other sustained real-world work. It doesn’t have a final exam. It has a record of what the student actually shipped.
The year-by-year curriculum I’ve described is the Essentials layer. Mastery and Excellence build alongside it. A student who takes all three seriously finishes with all three.
What this walkthrough doesn’t tell you
I want to be direct about one thing this post can’t answer.
The curriculum is what’s available. What a student actually builds depends on what they put in. DOJO runs six days a week, but only if the student shows up. The Internship or Project modules run from Semester 3 through Semester 8, but only if the student treats each one as a real piece of work rather than a completion target. The Final Capstone is as strong as the question the student chose to pursue and the execution they committed to.
As of March 2026, 82.40% of Kalvium’s first graduating batch were placed before completing their degree, with a median salary of ₹16.5 LPA. That number sits at the far end of the arc I’ve described. It doesn’t appear unless the four years actually happened.
For the learning science behind why this build arc produces better capability than a lecture-first alternative, why most engineering programmes don’t produce engineers makes the argument with the named research. The complete guide for families, covering the nine partner universities for Admission Year 2026-27, the KNET admissions process, and the full programme structure, is at what Kalvium actually is.
Arvind is Head of Programme Design and Delivery at Kalvium. He writes about the cognitive science of learning and how it shows up in the design of a live programme, grounded in named research and honest about where the evidence is still uncertain. Read more from Arvind or browse the B.Tech category.