Building in Tech: How to Evaluate Candidates using AI In the Hiring Process

AI is already part of how developers work today, and naturally, it’s becoming part of how we evaluate them, too. The question is no longer whether candidates use AI, but how they use it and what that reveals about their thinking, experience, and approach to problem-solving.

We asked several of our team leaders at T4 how they approach AI in the hiring process today, and where they draw the line.

TL;DR: AI in Developer Hiring Is No Longer About “If” But “How.”

Team leaders (and companies) aren’t evaluating AI usage itself, but the thinking behind it.

Key principles:

• AI is a tool - candidates must understand what they deliver.
• The focus shifts from coding to thinking, decision-making, and explaining solutions.
• Transparency, ownership, and respecting rules are essential.

By seniority:

• Junior: AI as a learning tool → understanding matters, not copy-paste
• Medior: AI as a collaborator → must critically evaluate and make decisions
• Senior: AI as a productivity tool → solution design must be their own

Emerging trend:

• Seniors use AI to scale their expertise across teams (tools, prompts, workflows)

Bottom line:

👉 It’s not about whether you use AI

👉 It’s about how you think, decide, and take ownership

AI Is a Tool - But Context Matters

For Jakub Polák, Mobile Development Team Leader, the key factor is seniority.

“For juniors, I value solutions created without AI more, because I want to see their own thinking and fundamentals. For seniors, using AI is completely fine, even expected in areas where it can significantly speed up the work.”

At the same time, he highlights a critical baseline:

“Regardless of seniority, the candidate must understand what they deliver and be able to explain both the solution and the thinking behind it.”

From Coding to Thinking

According to Dušan Drábik, Frontend Team Leader, AI is shifting the focus of what we should prioritize in interviews.

“We can’t avoid it, so we have to take it into account. We need to focus more on understanding concepts rather than programming itself.”

In practice, this means evaluating not just output, but reasoning, how candidates approach problems, make decisions, and adapt solutions.

Clear Boundaries: Transparency, Ownership, and Responsibility

Martin Jankovič, Backend Team Leader, outlines a few core principles when it comes to AI usage:

• Transparency: Candidates should openly say if and how they used AI. Hidden usage is a red flag.
• Ownership: It must be clear what the candidate’s own thinking is versus AI-generated output.
• Respecting rules: If AI is explicitly not allowed in an assignment, that boundary matters.
• Security awareness: Candidates should understand that prompts and data may be shared with third-party systems.

What to Expect at Each Seniority Level

The expectations around AI vary significantly with experience.

Junior

AI can be a learning tool - helping with syntax, small improvements, or understanding how to approach a problem.

What matters most:

• Ability to understand the task
• Breaking it down into steps
• Reviewing and validating the output

If a candidate simply copies AI output without understanding it, that’s where it crosses the line.

Medior

At this level, AI becomes more of a collaborator.

Candidates are expected to:

• Use AI for implementation support (boilerplate, refactoring, structure)
• Critically evaluate and adapt suggestions
• Make their own decisions about what to build and why

The key shift: AI can help write the code, but not define the solution.

Senior

For senior candidates, AI is a productivity tool, not a decision-maker.

They should:

• Design the solution independently
• Use AI to speed up execution or explore alternatives
• Clearly explain trade-offs, risks, and architectural decisions

The line is crossed when the core design, structure, or decisions come from AI – and the candidate can’t justify or adapt them.

A Shift in Team Dynamics

One emerging trend we’re starting to see:

Senior engineers increasingly create tools, prompts, or workflows that enable more junior developers to move faster -effectively “scaling down” their experience across the team.

This changes not only how teams work, but also how we evaluate talent:

We’re no longer just hiring for individual output, but for how someone uses and shapes tools around them.

So What Are We Really Evaluating?

Across all perspectives, one theme stands out:

AI doesn’t replace the need for skill - it changes what skill looks like.

Today, strong candidates are not defined by whether they use AI, but by:

• how they think
• how they make decisions
• how they take ownership of their work

Because in the end, AI can generate code, but it can’t take responsibility for it.