Over the last month, I’ve given multiple internship interviews and completed several technical assessments as part of internship hiring processes.
Obviously, the main goal was to get an internship.
But after going through these interviews and assessments, I realized that the process itself was teaching me a lot.
It improved the way I communicate, increased my confidence, made me much more comfortable with interviews, and also changed the way I use AI coding agents.
So I thought I’d share some of the things I learned from the experience.
My Communication Got Better
One of the biggest changes I noticed was in my communication.
Knowing something and being able to explain it clearly are two different skills.
During interviews, I had to explain my projects, the technologies I used, why I made certain technical decisions, how different parts of my projects worked, and how I would approach different problems.
When you build something yourself, a lot of this exists naturally in your head.
But during an interview, you have to turn those thoughts into a clear explanation that another person can understand.
After giving multiple interviews, I became much more comfortable doing that.
I can explain my projects and technical decisions more clearly now, and I’m also more comfortable having technical conversations in general.
My Confidence Increased
Another major change was confidence.
Earlier, interviews naturally came with some fear.
What if they ask something I don't know?
What if I can't answer a question?
What if something goes wrong?
After going through multiple internship interviews, that fear reduced a lot.
You eventually realize that you don't need to know the answer to every single question.
Sometimes you know the answer immediately.
Sometimes you need to think.
Sometimes you have to reason your way through the problem.
And sometimes you simply don't know something.
Going through that experience repeatedly made interviews feel much more normal to me.
Instead of constantly worrying about what might be asked, I can focus more on the actual conversation and the problem in front of me.
AI Coding Skills Are Becoming More Relevant
Another interesting thing I noticed during some of these internship interviews and assessments was the importance of AI coding tools.
I came across questions about tools like Claude Code and Codex and about using AI agents while coding.
That made one thing very clear to me:
Knowing how to use AI for coding is becoming a useful development skill.
But by "using AI," I don't mean just opening an agent and saying:
Build this feature.
The important part is knowing how to actually work with the agent.
For example:
- Giving it enough context about the codebase
- Clearly explaining what you want to achieve
- Breaking larger problems into manageable tasks
- Understanding the code it generates
- Reviewing its changes instead of blindly accepting them
- Finding issues in AI-generated code
- Recognizing when the agent is going in the wrong direction
- Giving it better direction when it gets stuck
- Testing and verifying the final implementation
These internship assessments gave me more opportunities to work this way, especially when I had to complete something within a limited amount of time.
Agentic Coding Isn't About Letting AI Do Everything
This is probably one of the most important things I learned from the whole experience.
Using an AI coding agent doesn't mean giving it a task and accepting whatever it produces.
AI agents can generate a lot of code quickly.
But they can also be wrong.
Sometimes an agent fixes one problem and creates another.
Sometimes it misunderstands the architecture.
Sometimes it changes something that didn't need to be changed.
And sometimes it gets stuck repeating almost the same solution again and again.
That's where your own technical understanding becomes important.
You need to understand what the agent changed and why.
When something doesn't work, you need to inspect the generated code, understand the error, identify where the problem actually is, and then either fix it yourself or give the agent better direction.
Over the last month, I became much better at recognizing these kinds of issues in AI-generated code.
I also became better at knowing when to let the agent continue working and when I should take control.
The Way I Use AI Coding Agents Got Better
Technical assessments usually have one thing that normal personal projects don't always have:
A strict deadline.
You might have only a few hours or a couple of days to understand the requirements, plan the solution, implement everything, test it, fix issues, and submit it.
That changes how you use AI.
You can't spend hours going back and forth with an agent while it keeps trying the wrong approach.
I learned to be more intentional about how I use coding agents.
Instead of expecting the agent to handle everything, I use it as part of my development workflow.
I decide what I want to build.
I understand the architecture.
I give the agent the necessary context.
I review what it produces.
If the generated code has issues, I identify them.
If the agent gets stuck, I change the approach or step in myself.
And before considering something complete, I verify that it actually works.
For me, becoming better at agentic coding isn't about getting AI to write more code.
It's about becoming better at working with AI while still staying in control of the code.
Working Under Time Limits Taught Me a Lot
The assessments also gave me experience building under pressure.
When you're working on your own project, it's easy to say:
"I'll finish this tomorrow."
An assessment doesn't always give you that option.
There is a deadline.
That means you have to understand the requirements quickly, decide what actually matters, manage your time, build the solution, test it, and submit it.
That experience helped me become more comfortable building projects within strict time limits.
It also forced me to think more carefully about where AI agents could actually save time and where spending time reviewing their output was more important.
The Biggest Thing I Got From These Interviews
Not every internship interview turns into an offer.
Not every technical assessment leads to the next round.
But I don't think that makes the experience useless.
Over the last month, these interviews and assessments have given me much more than just opportunities to apply for internships.
My communication improved.
My confidence increased.
My fear of interviews reduced significantly.
I became better at identifying problems in AI-generated code.
I became better at using AI coding agents effectively.
And I became more comfortable building projects under strict deadlines.
Most importantly, interviews don't feel like some huge scary event anymore.
They're conversations.
They're technical discussions.
They're opportunities to explain what I've built, solve problems, and experience the kind of situations that developers deal with in the real world.
Of course, I still want the internship at the end of the process.
But even before getting that final result, the process itself has already taught me a lot.
Top comments (0)