DEVOPSdigest

The talent conversation isn't limited to HR and People teams. It's equally a conversation about process, operations, technology, and innovation. As the CTO of a software nearshoring company with over 4,000 software engineers, I understand tech talent as the foundation of our organization and the focus of nearly every client conversation.

We're in a moment of rapid transformation in how software developers approach their work. According to our Dev Barometer Q3 2025 findings, 65% of developers say they're worried about falling behind on AI skills, and they're taking matters into their own hands. They're saving, on average, over seven hours a week thanks to AI tools, and most are reinvesting that time into learning. They're not waiting for permission or a better timing to learn. They're teaching themselves new skills, diving into prompt engineering (44%), AI/ML specialization (45%), and learning how to use AI to boost productivity across the board.

But here's the disconnect: while developers charge ahead, most companies are still stuck in neutral. Only 15% of our project managers report formal upskilling programs at client companies, and even large enterprises are often figuring it out as they go. That's not an ideal scenario, but the reasons aren't surprising. Unclear ROI, limited resources (time, budget, staff), and a lack of in-house AI expertise make it difficult to shape meaningful upskilling programs.

AI Is Reshaping, Not Replacing, the Developer Role

It's been said before, but it's important to make the point: AI isn't replacing software engineers anytime soon. Tools don't replace a craftsperson; they enhance the craft. Photoshop didn't replace photographers. AI won't replace developers. Take Klarna's example. They let go of 700 people trusting AI automation, but after quality dropped, they're rehiring humans. AI automation isn't wrong, but rushed leadership decisions risk overlooking its impact on talent and customers.

According to our survey, 57% of developers say AI helps them move away from repetitive tasks and toward more fulfilling work like architecture, strategy, and problem-solving. Many are also branching into prompt engineering or AI/ML roles. Code is still the output, but the mindset is shifting from execution to orchestration, with developers now thinking in AI terms and rewriting the playbook every day.

The next generation of engineering excellence won't be measured by how well someone writes code but by how well they can design resilient systems, solve complex business problems, and work symbiotically with AI.

Organizational Stagnation Is a Real Risk

If you're leading a tech team today, big or small (but particularly the big ones), the threat isn't AI outpacing your developers. It's your company not keeping up with them. Developers are already enhancing the entire development lifecycle with AI, taking on more strategic responsibilities than many leaders had envisioned for them.

I've said this before: You don't want someone who was great five years ago. You want someone who's great now. But you can't expect people to stay great unless you give them the space and support to grow.

Right now, that support is missing in many organizations. The cost of that gap becomes clearer every day, and our project managers report that just 3% of the projects they work on have AI fully integrated into delivery workflows. Developers are moving fast, but the systems around them aren't evolving at the same pace.

In many companies, AI expertise still lives in the hands of a few motivated individuals who are self-taught, experimenting on their own. Meanwhile, the organization as a whole is still in the phase of early pilots, without a formal strategy or consistent resourcing. That disconnect is creating a gap that's hard to bridge without intentional effort. Closing it means shifting focus from what AI could do to what people need, namely, hands-on training and collaborative spaces that turn curiosity into shared capability.

Developer-Led Learning Is a Force Multiplier

Here's the good news: developers are already doing the hard part. They're learning prompt engineering, exploring AI/ML specializations, and figuring out how to integrate GenAI into their day-to-day workflows. For many, it's a personal investment that's quietly reshaping how their teams operate.

What they often lack isn't motivation, it's structure. And that's where leadership comes in. At BairesDev, we channel this momentum through what we call Circles, our internal Centers of Excellence. These are grassroots communities of practice in areas like AI, DevOps, and Data Engineering, led by volunteer engineers who step up to guide their peers. They're fueled by passionate engineers who want to stay at the frontier. This program explores new practices, standardizes effective ones, and brings their learnings back to the wider organization.

This learning model doesn't require a massive investment. What it requires is recognition, support, and purpose. If your team has developers already experimenting and sharing, build on it. Remember that voluntary contributions thrive when people are given the space to do it and when it aligns with their personal motivations, from self-development to community recognition.

Upskilling Is Also Cultural

As AI reshapes our work, it also reshapes what we value in technical talent. The takeaway is learning to work more adaptively, collaboratively, and system-awarely.

Yes, technical depth matters and always will. Developers need fluency in building cloud-native systems, working with data pipelines, and understanding model architecture. Those skills don't live in isolation. To benefit from AI in production, teams need to embrace DevOps maturity, operational awareness, and cross-functional fluency.

Simultaneously, soft skills emerge as non-negotiable, now critical for infrastructure in distributed teams. Communication, accountability, critical thinking, and collaboration are what hold AI-enabled systems together. AI may help generate solutions, but humans still define the context, make the judgment calls, and take responsibility for outcomes. AI models can't be held accountable for business failure.

At its core, upskilling creates a culture where growth is shared and where developers lift each other and the systems around them. That's how companies turn personal curiosity into team strength, and team strength into lasting capability.