He graduated with a first class. Top of his batch, the kind of transcript that makes relatives ask when the wedding is. He's sitting across from a technical interviewer at a Colombo software company, and the question on the screen is nothing like anything he's seen before: a messy, half-broken function, and a request to find out why it's failing.
He knows every algorithm in the syllabus. He can prove time complexity on a whiteboard without blinking. But this isn't a proof. It's someone else's code, written the way real code actually looks: inconsistent, undocumented, doing three things at once. He stares at it for a long moment, then starts explaining Big O notation, because that's the tool he has. The interviewer nods politely and moves on. He doesn't get the job.
Nobody in that room did anything wrong. He studied hard, followed the syllabus, did exactly what the system asked of him. That's precisely the problem this post is about.
This Isn't About Universities Being Bad
Let's be clear about what this argument is and isn't. It isn't that Sri Lankan universities are failing, that lecturers don't care, or that a degree has no value. Most lecturers are genuinely trying to prepare students well, often with limited resources and syllabi that take years to update. The argument here is structural, not personal: universities are optimised to produce a specific outcome, academic performance, and that outcome doesn't automatically translate into what a hiring manager at a tech company is looking for in 2026.
The better a student is at being a student, at memorising, at completing assignments to a marking scheme, at anticipating what an examiner wants to see, the more likely they are to walk into an interview having practiced the wrong game entirely. That's not a character flaw. It's what happens when the incentive structure rewards one skill set for four years and the job market rewards a different one.
A recent academic analysis of Sri Lanka's education system put this plainly: the country runs one of the most examination-centered systems in South Asia, where academic mobility from O/Levels through university is tightly linked to exam performance. The researchers noted that when progression is structured around high-stakes testing, students often shift their effort toward maximizing scores rather than building the applied competencies that a score is supposed to represent. Passing the test and being ready for the job start to quietly diverge.
What the Data Actually Shows
This isn't just a theory. It shows up in the numbers wherever people have bothered to measure it.
Sri Lanka's IT sector has grown fast, and industry projections from SLASSCOM point to a need for roughly 25,000 graduate-level IT hires annually to support the sector's export growth targets. That's a lot of open roles. And yet employers consistently report they can't fill them with the graduates coming out of the pipeline. Career guides tracking the local graduate market put the figure at 60% of employers reporting difficulty finding sufficiently skilled graduates, even in a market with plenty of degree holders actively job hunting.
Sri Lanka's Graduate Employment Reality by Sector
Notice what that chart is actually saying. IT graduates in Sri Lanka have one of the highest employment rates of any field, roughly 85 to 90%. That sounds like good news, and it partly is. But that same academic study found that even among graduates who do get hired, employers report they often need 6 to 12 months of on-the-job onboarding before those graduates reach independent productivity, despite arriving with strong academic records. Getting hired and being ready to contribute turned out to be two very different milestones.
What University Rewards vs. What Companies Actually Need
The gap becomes obvious the moment you put the two side by side.
| What University Education Typically Rewards | What Sri Lankan Tech Companies Actually Need from a Junior Hire |
|---|---|
| Memorising theory well enough to reproduce it under exam conditions | Applying theory to a problem that's never been explicitly taught |
| Solving problems with a known, single correct answer | Solving problems with ambiguous requirements and multiple workable solutions |
| Working alone against a marking scheme | Working inside a team, taking feedback, adjusting to someone else's codebase |
| Meeting a syllabus deadline set months in advance | Meeting a client deadline that shifts, gets renegotiated, or gets moved up |
| Writing code that compiles and passes the test case given | Writing code that survives contact with real, messy, undocumented systems |
| Being evaluated once, at the end, by an exam | Being evaluated continuously, in code reviews, standups, and retrospectives |
| Optimising for the correct answer the examiner wants | Communicating trade-offs to a non-technical stakeholder who has no marking scheme in mind |
None of the left column is worthless. Data structures, algorithms, and the discipline of exam preparation are real and useful. But if that's the only column a student trains in for four years, the right column shows up for the first time in a job interview, which is a genuinely unfair place to encounter it for the first time.
A degree tells an employer you can perform under a system with clear rules. It says almost nothing about whether you can perform when the rules are unclear, which is most of what the job actually is.
In Fairness: Degrees Still Matter, and Here's Where
It would be dishonest to argue that degrees are pointless, because in several very real ways, they're not.
Visa and immigration pathways in most countries still require a formal degree as a baseline eligibility criterion, and no amount of practical portfolio work substitutes for that piece of paper on an application form. Large multinational corporations and public sector employers, particularly banks and government-linked institutions, often use a degree as a hard filter before a resume is even read by a human, regardless of how the market talks about "skills-based hiring." And there's a deeper point too: the theoretical grounding a degree provides, understanding why an algorithm behaves the way it does, why a database index matters, why concurrency is hard, gives graduates a foundation that pure bootcamp-style training sometimes skips past too quickly. Companies building genuinely complex systems value engineers who understand the "why," not just the "how."
The honest conclusion isn't "skip university." It's that a degree is necessary in specific, identifiable situations, and increasingly insufficient on its own everywhere else.
What a Hiring Manager Actually Sees
Picture two candidates for the same junior developer role. Both hold the same degree from the same year. Their transcripts look nearly identical.
Candidate A has a GPA and a final year project that was scoped, assigned, and marked by a supervisor according to a rubric. That's the entirety of their technical evidence. Candidate B has the same degree, plus two or three real projects built outside the classroom, a GitHub history that shows commits over months rather than a single upload before a deadline, and maybe a few months of internship or freelance work where they had to explain a technical decision to someone who wasn't grading them on it.
The hiring manager isn't comparing intelligence. Both candidates are probably comparably capable. What they're actually comparing is evidence of the right column in that table above: has this person ever had to work through ambiguity, defend a decision, fix something they didn't originally build, or ship something under a deadline that wasn't handed to them by an institution. Candidate A is a well-documented unknown. Candidate B is a known quantity. In a market where, as the same academic research notes, graduates typically need 6 to 12 months to reach independent productivity, a hiring manager who can shave even a few of those months off by hiring someone with real practical exposure will do it every time.
This is really the whole story. It's not that Candidate A is less talented. It's that Candidate A's four years of evidence answer a question the interview isn't asking.
Closing the Gap Without Abandoning the Degree
None of this means throwing away four years of study. It means recognising that a degree answers one question and the job market is asking several others, and treating the gap between them as something to actively close rather than something to discover in an interview room.
That closing happens through the things that generate the right column in the table: real projects with genuine constraints, not graded assignments with a rubric attached. Exposure to how an actual team works, where code gets reviewed, deadlines get renegotiated, and decisions get explained to people who don't already know the answer. Verified, demonstrable proof of practical skill that a hiring manager can actually evaluate, rather than a GPA that tells them how well someone performs under exam conditions.
Structured, project-based training exists specifically to build that second column while a degree is still building the first. Neither replaces the other. The graduates who walk into interviews and don't freeze on the practical question are almost never the ones who studied less. They're the ones who spent part of their four years doing something the syllabus never graded.
What IT Employers Say They Struggle to Find in Graduates
That 60% gap isn't a mystery to be solved by another exam. It's solved by graduates walking in with something to show beyond a transcript, and by an industry willing to look past the GPA long enough to see it.
Where This Leaves You
If you finished university and felt underprepared in your first technical interview, that wasn't a verdict on your intelligence or your work ethic. It was the predictable outcome of a four-year system optimised to reward one very specific skill, tested against a job market that quietly needs several others.
The fix isn't resentment toward the degree, and it isn't pretending it doesn't matter. It's building the practical, evidence-based half of your profile deliberately, on purpose, alongside the academic half, instead of hoping the classroom eventually covers it. Nobody hands you that second half automatically. But it's entirely learnable, and the graduates who go looking for it tend to stop freezing in interviews.
