The Rise of AI: Keep the Human Pilots in the Cockpit

I learnt to write my first if-statement back in 2003. Two decades later the same building blocks—algorithms, branches, loops, functions—still power everything from e-commerce checkouts to fintech ledgers and mining dashboards. Sometimes, it feels like my fingers on autopilot because I’ve done it over and over again and the patterns are already in muscle memory. But, sometimes it required deeper thinking because I needed to consider specific business requirements from multiple angles. Often, I needed to stop and juggle trade-offs, sacrificing speeds over security, accepting slower writes for stronger consistency, or lowering uptime to minimize cloud spend. And it was just another “normal days” in my daily work life to have Stack Overflow and Reddit glow on my second monitor for those moments.

I’ve watched plenty of tech waves rise and fade, and it’s clear AI is next in line. Copilot-style tools now crank out in seconds what once took me hours—often with uncanny accuracy. No wonder many people glorify “vibe coding” lately. It’s all about getting things done and being productive. Seasoned engineers push back, saying that it’s not an engineering, because the focus is only to make the code runs without even considering the quality or the future maintainability.

Honestly, I’m torn. For the last two years, ChatGPT and Claude have lived on my second screen, and Copilot is always enabled on VS Code, and yep, they speed me up. But yet, I can still proudly say that I am the “pilot” in this flight. I remember one day in my early professional career as Junior Dev, I got flagged from my senior during code review. I didn’t feel that I’ve done something wrong because my code ran fine. But, it was the first time I heard this term: “spaghetti code”. The code might be functional, but it’s like a time bomb that can explode at any time because of the accumulated tech debts. Maybe that’s how I look at vibe coding today: slick and fast, but hiding future pain.

So here’s my note to fellow CTOs and IT directors: think twice before you slash headcount and bet the house on AI. Of course there will be thousands of reasons to do “efficiency” in current economical situations, but dumping good engineers over AI is nothing but a liability. Ensuring high productivity was your job before AI and it still is. Keep the pilots; let the bots copilot.

And as always, stick around—this ride is bumpy, but it doesn’t have to be a crash.