Introduction: When Reality Hits Different
My name is Namaneetha, and I recently landed my first internship as an SDE-1. Here's the thing though—software development wasn't even my dream. My passion lies in digital forensics and incident response. But when you're starting out, you take what comes your way, right? This was the first company that said yes to my resume, so I said yes back.
My first assignment seemed exciting: develop a Chrome extension for URL analysis using machine learning and rule-based mechanisms. Spoiler alert—it didn't go as smoothly as I hoped. This is the story of what happened when I relied too much on AI, lost control of my own project, and learned some hard lessons about what it really means to be a developer.
The Assignment: Building a URL Analysis Chrome Extension
The project requirements were clear: create a Chrome extension that could analyze URLs for potential threats using both ML models and rule-based detection. Following the guidelines, I dove in and managed to build something that achieved 92% accuracy. I even finished before the deadline. I felt pretty good about myself.
Then came the curveball.
The team wanted additional features. They handed me a completely different dataset to retrain the ML model. And that's when my carefully constructed house of cards started to collapse.
Where Everything Went Wrong: The AI Dependency Trap
Here's the uncomfortable truth I need to admit: I didn't really build that first version. Not in the way I should have.
When faced with this complex project, I leaned heavily—too heavily—on AI assistants like DeepSeek and Grok. Instead of struggling through problems and actually learning, I became more of a project coordinator. I'd take code from AI, align the pieces, make sure everything ran, and call it done. The project worked, but I had no idea why it worked.
The moment of reckoning came when I tried to add those new features.
I stared at my own code like it was written in a foreign language. The structure made no sense to me. I couldn't see how the pieces fit together. I couldn't modify anything without breaking everything else. That's when it hit me: I had over-relied on AI to the point where I didn't understand my own project.
The Scramble: Two Days of Panic and Patches
Realizing my mistake, I tried to start over and rebuild the project properly. But the deadline was already looming. I was caught between wanting to do it right and needing to deliver something.
Those final two days were brutal. Barely any sleep. Skipped meals. Just me, my laptop, and a growing sense of inadequacy as I frantically patched things together with more AI assistance because I was out of time to learn properly.
I submitted the project. It was done. But was I satisfied? Not even close.
The Emotional Aftermath: Questioning Everything
After completing the project, I felt completely discouraged. Instead of feeling accomplished, I felt like a fraud. The questions started spinning in my head:
- Do I even have talent for this?
- Did I choose the wrong career path?
- Am I cut out for tech at all?
It's a weird feeling—completing a project but feeling worse about yourself than when you started. That assignment didn't just teach me about code; it taught me about the difference between finishing something and truly understanding it.
The Lessons I'm Taking Forward
Looking back, here's what I learned from this mess:
On Using AI
AI is a powerful tool, not a replacement for learning. There's a huge difference between using AI to help you understand a concept and using AI to avoid understanding it altogether. I crossed that line, and it cost me.
On Building Skills
You can't skip the struggle. The frustration of debugging, the confusion of trying to understand how things work, the slow process of building something yourself—that's not wasted time. That's where the actual learning happens.
On Self-Awareness
Recognizing your limits is the first step to growing. My mistake wasn't accepting a project that was challenging. My mistake was pretending I understood something I didn't, then letting AI mask that gap instead of filling it.
Moving Forward: My Commitment to Real Growth
This experience shook me, but it also gave me clarity. I've made some concrete changes:
1. Structured Learning Schedule
I'm dedicating time every day to actually learn programming concepts from the ground up. Not just copying solutions—understanding them.
2. Balanced AI Usage
I still use AI assistants, but now as learning tools. I ask them to explain, not just to solve. I try to implement solutions myself first, then use AI to check my understanding.
3. Daily Documentation
I'm posting my daily learnings regularly. Partly for accountability, partly to help others who might be struggling with similar challenges.
My heart is still in digital forensics and incident response. But I now understand that every technical experience—even the difficult, humbling ones—makes me better equipped for that future.
To Anyone Starting Out
If you're reading this and you've over-relied on AI, made mistakes on your first project, or felt like an impostor—you're not alone. We all stumble. What matters is whether you choose to learn from it or let it define you.
I'm choosing growth over giving up.
This isn't the story of a perfect first project. It's the story of a messy one, and the commitment to do better next time.
And I'll be sharing every step of that journey right here on DEV.
What about you? Have you ever found yourself over-relying on tools and realized you'd lost understanding of your own work? I'd love to hear your experiences in the comments below. 👇
Top comments (0)