Generative AI has shifted the hiring landscape, sparking ongoing debate among employers and job seekers about whether a traditional computer science degree still holds value. Though degrees remain a powerful signal of structured, algorithmic thinking in the traditional hiring process, the skills gap between credentialed graduates and self-taught, AI-assisted developers is narrowing.
As technical vetting pivots away from rigid gatekeeping toward practical, skills-based assessment, how should universities and developers evolve to ensure they’re not just building software but mastering the underlying systems that AI tools can't yet solve on their own?
How has the rise of generative AI changed your valuation of a formal computer science degree versus hands-on portfolio experience?
Bryan Wall, Senior Competency Leader, Software & Cloud Engineering at Experis, says “because of Gen AI, employers have now shifted to how we think about evaluating talent. Having a computer science degree still shows that humans can think algorithmically, understand systems at a foundational level, and process complex problems that AI tools can't always solve on their own. That being said, candidates without traditional degrees are building these sophisticated portfolios by using AI-assisted development, and those portfolios are increasingly hard to ignore.”
“Degrees are still valuable but AI has changed what that value is and how much weight it should carry in hiring,” Will Steward, CEO & Co-founder, The SaaS Jobs, tells Dice. “A good computer science or engineering degree still tends to deliver structured fundamentals that are hard to patch together ad hoc. Those foundations matter even more when AI can generate code quickly, because the differentiator shifts toward evaluating trade-offs, spotting subtle failure modes, and designing systems that are reliable, testable, and secure at scale. AI reduces the time spent typing routine code, but it doesn't remove the need to understand what the code is doing or how it behaves in production.
“That said, degrees no longer need to function as a universal hiring gate. AI-assisted workflows have made it easier for capable candidates without a traditional background to build credible evidence of ability through shipped projects, open-source contributions, and work samples that demonstrate judgment, maintainability, and testing habits. This has accelerated the move toward skills-based screening, where interviews lean more on practical assessments and deeper technical discussion than on credentials alone. The direction of travel is reinforced by how common AI tooling has become in day-to-day development.”
What foundational, systemic concepts do university degrees provide that self-taught, AI-assisted developers frequently lack?
“A degree provides a framework for thinking about systems, not just solution,” claims Mark Friend, Company Director at Classroom365, a UK-recognized leader in school IT support. “A successful self-taught developer will typically focus on fixing problems faced at the moment without understanding the overall system or the reasons that caused it. Self-taught developers learn through trial and error and although they have a great understanding of how to build, they lose out on the reasoning behind why a development process is done this way.
“Through education, you will learn how to conceptualize data structures, memory allocation, computational logic and more, which are all the essential structural building blocks of an entire system when it is under stress and unable to continue functioning correctly. Without these foundational concepts, development is left up to guesswork. If something goes wrong deep within a system, then a self-taught developer may not know where the issue lies.
“AI-assisted learning is compounding this problem instead of resolving it. AI provides developers with quick answers to their problems but does not give them the reasoning behind the manufacturing of those answers.”
Wal adds “generally, universities build programs that are meticulous by design. They're structured to ensure students develop a comprehensive understanding of computation from the ground up. Whether it's calculating math, or dissecting data structures and operating systems, this curriculum exists for a reason. It builds a mental model of how systems work beneath the surface. Students must understand the why behind the computation, not just the how, and that distinction matters enormously when problems get hard.”
Is a traditional four-year tech degree still a viable financial investment when AI is accelerating the obsolescence of hard skills?
“I believe that a traditional four-year technology degree is still a strong financial investment. The return isn't just in the credential itself, it's in the intellectual infrastructure that a rigorous program builds over four years of sustained, structured challenge. For most people, that environment is very difficult to replicate independently, and the market continues to recognize it,” Wall advises.
“The value of a four-year degree has shifted, and we need to recognize that. Not long ago, a computer science degree was the gateway to employment because it taught the hard skills employers needed. Today, AI tools are automating or augmenting many of those entry-level hard skills, from basic coding and debugging to data entry and documentation. As we found in our Q2 Tech Talent Report, when it comes to the actions employers are taking to overcome talent scarcity, we learned that 12% of employers are reducing or removing degree requirements while 29% are upskilling and reskilling current employees – including AI.”
“Yes!,” exclaims Yoan Ante, founder at Mathbuilders. “No matter how many hard skills become 'obsolete', someone always needs to know how the sausage is made. It has been proven time and time again that AI cannot understand or infer, it can only mimic extremely well. A four year degree at the right institution will provide deep insights into how systems interoperate. You can't get that with surface level coding knowledge. If something breaks and the AI can't fix it, which still happens daily, someone needs to come in and figure out why. Those people will out earn the ones that can produce code fast but can't solve problems when things break.”
How are you adjusting your technical vetting processes to verify deep engineering aptitude in candidates who do not possess a formal degree?
“My approach to technical vetting remains unchanged regardless of whether a candidate holds a formal degree or not, says Wall. “I focus on having an engineer-to-engineer conversation about the systems they have built, the challenges they have faced, the architectural decisions they've inherited, and how they would change those decisions in hindsight. A formal degree can provide a strong foundation, but the most valuable learning occurs on the job, where engineers are required to solve complex problems that can appear to have no clear answer.”
What critical structural changes must university computer science curricula implement immediately to ensure graduates remain competitive against automated tools?
“For too long, some universities leaned on the paper itself as the value proposition. The degree was the product,” Ante points out.
“That has to shift toward real outcomes. Can the student build? Can they debug? Can they read messy code? Can they work on a real project with real constraints?
The curriculum has to produce at least as much as it teaches theory. Theory still matters, but if it never turns into working software, it is not enough anymore.
Bootcamps were the first real pressure test for universities. I think AI is the next change agent. Both are forcing schools to prove they are not just filling students’ heads with information, but preparing them to actually do the job.”
“University computer science departments need to face the fact that teaching students to write code is no longer sufficient,” Wall points out. “Instead, they must teach students to think at a level that automated tools fundamentally cannot. Programs must also close the gap between academic work and professional reality by building in meaningful capstone experiences, industry partnerships, and project-based learning that mirrors the ambiguity and complexity graduates will face on the job.
“Having soft skills like communication, collaboration, and translating technical decisions for non-technical stakeholders, must be treated as core assets. While clients who have recent graduates who stand out aren't necessarily the strongest coders, they're the ones who can frame a problem clearly and lead a room through a difficult technical decision. Universities that can restructure these types of developments will produce graduates who aren't competing against automated tools. They'll be the ones directing them.”
An evolving hiring landscape doesn't signal the obsolescence of the computer science degree, but rather a maturation of what is valued in technical talent assessments. While AI can drastically accelerate code creation, it cannot replicate the deep architectural intuition required to navigate complex system failures or to design for long-term scalability. In this new era, those who learn to view AI as a powerful tool rather than a replacement will find themselves leading the industry rather than competing against the automation it produces.
