Being a junior in the age of AI
The interviewer looked at me, let go of the mouse, and said: “You’re not cut out for this.”
I’d been in the interview for five minutes.
Five minutes. No VS Code, no documentation, nothing. Just me, the browser DevTools, jQuery (yes, jQuery, in July 2025), and a Head of Tech who decided, faster than it takes to make a coffee, that I didn’t belong in this field.
I walked out feeling lost. That afternoon I couldn’t stop wondering if I’d made the right choice. If those two years of vocational training, waking up early to deliver packages in the morning and going to class in the afternoon, had been worth anything. If the market was telling me something I didn’t want to hear.
Spoiler: it wasn’t. Or rather, it was, but not what I thought at the time.
Ten months later, I’m part of a team, working on real projects with Angular, and writing this hoping that what I learned the hard way helps someone who’s where I was.
The easy diagnosis won’t help you
There’s a very convenient narrative going around LinkedIn lately. You know it. “Companies now ask for two years of experience for junior roles.” “AI is eating entry-level jobs.” “The market is broken.”
And yes, some of that is true. The junior market is more competitive than it was three years ago. The bar is higher. That’s real.
But there’s a trap in stopping there: it turns you into a victim of something you can’t control. And that mindset freezes you and hurts your job search. I’ve seen strong profiles stuck in that loop, complaining no one gives them a chance. But maybe you’re not asking yourself this: am I actually competing well in this market?
The usual portfolio doesn’t stand out anymore
Here’s something people don’t say out loud: if your portfolio has a clone of a famous app, a task manager, and a weather app, the recruiter has been seeing the exact same thing since 9 a.m.
Those projects are great for learning fundamentals. Perfect for that. But they don’t make you stand out—because everyone else has the same ones.
The better question is: what real problem in my life could I solve with code?
It doesn’t need to be a startup. It doesn’t need to be perfect. It needs to be yours, with a story you can tell. “I built this because I ran into this problem, tried this, failed here, fixed it like this.” That’s what sticks in the mind of the person interviewing you.
And about AI: use it. Please, use it. But to learn, not to copy-paste without understanding what’s going on. If you ask Copilot, Claude Code, Codex, etc. to write a function and you drop it into your project without being able to explain it line by line, it shows. And in a technical interview, it really shows.
AI is an incredible tool to speed up learning if you use it right. To explore tech you’d otherwise take weeks to grasp. To debug when you’ve been staring at the same error for two hours. To understand patterns tutorials don’t fully explain. Use it to ask the questions you’d be embarrassed to ask out loud. Use it to get the same concept explained five different ways until one clicks. That’s learning with AI. That multiplies.
Also use it to build that project you’ve had in your head for ages but thought, “I can’t do this alone.” Now, in the era of AI agents, the possibilities are huge. Take advantage of it.
What doesn’t multiply is using it as a shortcut to avoid thinking. Blind copy-paste won’t save you in an interview and it won’t save you in your first month on the job when someone asks why you did something a certain way. AI won’t take your job, but someone who uses it better than you might.
Make mistakes. Then make them again. Understand why. That’s the process. There’s no shortcut.
How I looked for a job and what I was doing wrong
Back to July. After that terrible interview, I was pissed. But it wore off quickly, because deep down I knew that company and I weren’t a match. If someone makes you write jQuery without documentation in DevTools in 2025, that’s not where you’re going to grow.
What did make me think was how I was approaching the job search.
I was applying to too many roles with the same CV. Sending applications in bulk, hoping volume would make up for lack of precision. I wasn’t really reading job descriptions, wasn’t adapting anything, just shooting and hoping. And summer is a bad time: the industry slows down, there’s less movement, hiring decisions get pushed to September. So instead of banging my head against the same wall, I made a call: pause, use that downtime, and be ready for September.
I started properly researching how hiring processes work. What recruiters look for in a CV and how long they spend on each one. What formats work and what don’t. How to read between the lines of a job post to understand what they actually want beyond the listed requirements. I rebuilt my CV from scratch. Not a quick polish, line by line, intentional. And most importantly, I became selective. Instead of applying to anything labeled “junior frontend,” I chose roles where I actually fit and tailored each application.
At the end of August, I saw a junior frontend Angular role that matched what I wanted. I optimized my CV specifically for that role, applied and three days later, I was hired. Initial call, technical interview, confirmation call.
Three days.
After months. Three days.
It wasn’t luck. It was doing the work I hadn’t done before.
What matters in interviews and isn’t in any guide
The technical part matters. Obviously. But there’s something just as important, if not more, that almost no one talks about: how you behave when you don’t know something.
When a company hires a junior, they know they’ll be investing time in that person for months. The question the interviewer is asking, even if they don’t say it out loud, is: do I actually want to have this person next to me while this person learns? Does this person listen? Does this person have the judgment to know when to ask and when to just dive in?
You don’t show that by memorizing the difference between var, let, and const. You show it in how you react when you get stuck. If, when you don’t know something, you become defensive or go blank, or if you say: “I don’t fully know this, but the approach reminds me of X and I think I could get there in this amount of time.” The second answer says more than memorizing anything ever could.
I went to interviews where I performed better technically than in others and didn’t get through. And I passed interviews where I got stuck at some point. The difference was how I handled those uncomfortable moments.
For anyone stuck in the loop
I won’t tell you to “believe in yourself.” That alone doesn’t work. Here’s something more concrete.
If you’ve been applying for a while with no results, before concluding the market is broken, ask yourself: Are you applying everywhere without filtering or being selective and tailoring each application? Is your LinkedIn actually optimized, or did you set it up two years ago and forget it? Do you go to industry events and stay to network?
Do you have something to talk about beyond course, bootcamp, or degree projects?
Can you explain how you think, not just what you know?
Every rejection hurts. Every ignored application without feedback too. I get it. But if you start seeing each “no” as information instead of a verdict, you move differently. Someone told me I wasn’t cut out for this in five minutes. Today I’m part of the industry, writing this.
The market is tough. But people are getting in. And most of them don’t have the most impressive profiles. They’re the ones who understand how the game works, prepare well, and don’t quit when things don’t go their way.
That’s what I did. And I’m still learning every day, which is exactly what I want.
Want to know a little more about who wrote this article? 👇
Frontend Software Engineer at Softtek. In his free time, he trains middle-distance athletics at his local club and recently took up road cycling. He enjoys building products that solve real problems. He’s also a full-time dad, which is also a job.