Gartner predicts that by 2029, 60% of organisations will adopt smaller software engineering teams at scale, compared with 15% in 2026. Meanwhile, a 2026 Microsoft Research study involving 331 software developers found that AI was enabling some professionals to take on tasks previously beyond their expertise.

This creates an important question: Does being able to complete a task mean we have developed the expertise to evaluate it?

From curiosity to ownership

For Karolina Didikaitė, professional growth has never been about following a traditional career ladder. After nearly five years in biotechnology, she moved into IT and became a Scrum Master. At Adroiti, her curiosity led her towards technical quality assurance (QA), where she is now developing new competencies while gradually transitioning from her previous responsibilities.

"I don't see myself staying in one field throughout my career. I want to keep learning and looking at things from different perspectives. I don't really see myself following a vertical career path. I'm much more interested in growing horizontally," Karolina explains.

Initially, she expected to learn software testing in her spare time. But her team encouraged her to go further and take responsibility for testing a specific part of the product: "I thought I'd just learn whenever I had some free time. But then my colleagues said: there's a part of the product you can take ownership of. And my first reaction was: how am I supposed to do this on my own?"

The encouragement came with practical support. Karolina received opportunities to learn through real work, while her colleague Berta took an active role in guiding and coordinating her development.

"The support from my colleagues was incredibly important. You join a company for one role and then say you'd like to learn something different. In some places, you'd simply be told to focus on your existing responsibilities. Here, I received genuine support and opportunities to grow."

For Karolina, taking ownership became a way to turn curiosity into practical experience.

AI makes learning faster. But does it make us better?

AI arrived at an important moment in Karolina's transition. It helped her navigate technical documentation, investigate previous product decisions and explore unfamiliar concepts without having to rely on colleagues for every answer.

"Before, when I was learning something new, there was often this uncertainty. You sometimes pretend you understand something, even when you're not completely sure. Now I can ask questions, explore different explanations and really work things out."

For example, while working on software testing, Karolina used AI to investigate an inconsistency between an existing test case and a previous product decision. By tracing earlier tasks and documentation, she helped identify that the test case had not been updated. The experience showed how AI can accelerate technical investigation, while understanding the product context remains essential to interpreting the findings.

But Karolina also recognizes the other side of AI-powered productivity.

"Now you don't have to remember or search for nearly as much information as before. Everything is available so quickly. Sometimes you don't even check the sources. When you're working on several things at once and constantly trying to move faster, you can miss something important. Sometimes you don't even remember what you were working on."

These observations reflect a broader challenge for technology professionals: AI can accelerate execution without necessarily developing the understanding needed to evaluate the results.

For Karolina, the difference has been the combination of AI tools, real responsibilities and experienced colleagues who help her make sense of what she is learning.

"I can explore things independently, get answers and then discuss the suggestions with my colleagues. When you ask good questions, you can get really useful results."

Why human mentorship still matters

Mentorship helps software professionals develop the judgment that AI tools alone cannot guarantee. While AI can explain technical concepts, generate solutions and accelerate learning, experienced engineers help validate decisions, understand product context and recognize potential risks. Combined with real project ownership, mentorship turns AI-assisted productivity into practical expertise.

Karolina's experience illustrates why this matters.

AI helped her learn faster, but it was the team's trust, Berta's guidance and the opportunity to take responsibility for real product work that made her transition possible: "It was really a combination of things coming together at the right time. The support, the opportunity to learn and the AI tools. I couldn't have hoped for better timing."

Her story offers three lessons for technology professionals and their employers.

• Productivity is not the same as expertise. Being able to deliver more with AI doesn't automatically mean understanding more.
• Mentorship remains essential to meaningful development. Experienced colleagues help turn information into judgment and practical knowledge.
• Ownership creates growth when responsibility comes with support. Giving people the opportunity to step beyond their existing roles matters, but so does helping them develop the competencies to succeed.

At Adroiti Technologies, our approach to professional growth combines continuous learning, engineering mentorship and ownership of real product challenges. Karolina's experience illustrates how an AI-first mindset works in practice: technology accelerates learning, experienced colleagues help develop judgment, and trust creates opportunities to take on greater responsibility.

Because in software engineering, the real measure of growth is not simply what AI helps us produce. It's what we understand, can evaluate and are ready to take ownership of.

Follow our latest insights on Adroiti Linkedin!

Adroiti is a Lithuanian technology company building AI-powered systems and running senior engineering teams that deliver real, production-ready results.