Should I Be Concerned About AI as a Software Engineer?
YES. You should be very concerned.
I know it may be uncomfortable to hear, but this is the truth. There’s no hiding from it or downplaying it just because it doesn’t feel relatable. This is one of the biggest pivots you'll ever make in your software engineering career.
This is your Kodak moment.
⚠️ Remember Kodak?
Kodak invented the first digital camera in 1975.
But they ignored it.
They feared it would kill their cash cow—film.
They rested on their legacy, while competitors surged ahead.
That hesitation cost them everything.
You are Kodak.
AI is your digital camera.
You’ve played with AI tools—improved your writing, polished your emails—but that’s not enough anymore.
You need a production line now. You need systems, workflows, and full-on AI adoption.
💡 Your 10x Opportunity
I know this might ruffle some feathers.
But the truth is: AI levels the playing field.
It’s not about whether you can write low-level memory allocators. It’s about whether you can deliver more value, faster.
CRUD APIs? Should take < 1 hour.
Unit & integration tests? Auto-generated.
CI/CD? AI-assisted, efficient.
You can—and should—be a 10x output engineer because of AI.
🎮 Vibe Coding = Real Innovation
One of the most exciting recent trends on X was “vibe coding”—building games and tools using AI + Three.js in a matter of days.
Example:
Pieter Levels built a massively multiplayer flight simulator using Cursor + Claude.
Check it out → fly.pieter.com
My own project?
Driving on Mars — mars-4.vercel.app
It became a meditative sandbox for me.
We weren’t game developers. AI made us feel like we were.
🧼 “But That’s Not Real Coding…”
I hear it often:
“Using AI isn’t real coding.”
It. Doesn’t. Matter.
You ship. You solve problems.
Your tools help you scale yourself.
That’s engineering.
💼 In the Layoff Era: Do More With Less
Companies are laying off engineers.
Yet, those systems? Still running.
The ones left behind?
Now asked to do more with less.
AI is your only scalable solution.
It helps you ramp up, explore unknown codebases, and build faster than ever.
🔮 What’s Next?
Two camps are emerging:
AGI Camp (Altman, Musk, Zuck):
AGI in the next 12–18 months. Full superintelligence.
Realist Camp (LeCun, Meta AI):
We’ve hit a plateau. LLMs can’t reason. Scaling is costly.
Whatever you believe, one thing is true:
AI has already changed how we build software.
✅ What You Can Do Today
1. Try vibe coding
Build something small and weird. See how far AI can take it.
2. Learn system design
AI can code. But you must lead it.
3. Use .rules or .cursorrules
Teach AI to respect your codebase standards.
4. Prototype like crazy
Proof-of-concepts eliminate guesswork and accelerate buy-in.
5. Review with AI
Before sending code to your team, have AI test and critique it.
6. Always provide context
AI thrives on clarity. Think deeply, prompt thoroughly.
🧠 Final Thoughts
Having an AI-first mindset is not optional.
Writing prompts for ChatGPT isn’t enough anymore.
You need a process upgrade.
You need working agreements with yourself.
Start using tools like:
VS Code + Copilot
Cursor + Claude Sonnet 4
Gemini 2.5 (great for architectural support)
These tools can now scaffold full-stack apps.
They understand file structure, context, and intent.
You just need to ask the right way.
📣 Let’s Talk More
I’d love to hear how you’re adapting to AI in your career.
Let’s connect 👇