In the early ’90s, when I started my journey in IT, businesses frequently turned to spreadsheets as their first taste of automation. Tools like Lotus 1-2-3 and later Microsoft Excel empowered people to manage complex calculations, data entry, and reporting tasks.
Soon enough, enthusiastic users began writing code within spreadsheets, crafting sophisticated solutions that addressed their immediate business needs.
As businesses grew, their spreadsheets could no longer keep up, pushing them towards databases like dBase, FoxPro, and Microsoft Access. I recall seeing numerous books titled “Learn Microsoft Access in 21 Days,” reflecting people’s eagerness to quickly build tailored solutions. These efforts were practical, effective, and empowering—until they weren’t.
Eventually, these homegrown database applications reached their limits, and businesses turned to software specialists to build more robust, scalable solutions.
(If you prefer listening to reading)
Today, we’re witnessing a similar pattern with AI.
Non-developers are now leveraging AI tools to rapidly prototype and implement solutions for pressing problems. Just a few days ago, I used Google’s AI Studio to create two small applications quickly.
Sure, I could’ve written the code myself, but I simply needed my problem solved without the overhead of coding. Similarly, someone without a technical background could use AI tools to solve an immediate issue and possibly even launch a new business.
Yet, history shows us that these initial solutions rarely scale well. Eventually, businesses outgrow them, encountering roadblocks like:
- poor scalability,
- suboptimal architecture,
- or limitations in the AI models themselves.
That’s precisely when specialists—those who understand both technology and the underlying business problems—step in to refine or rebuild these solutions.
As software professionals, we have a valuable opportunity to pay close attention to how people use AI tools today. Just as we’ve learned to identify and resolve issues like redundant data in amateur databases, we should now learn the common pitfalls in AI-driven solutions.
When the time comes, we’ll be ready to offer precise, effective guidance.
I’ve often heard developers express anxiety about AI “taking their jobs.” But reflecting on my 30-year career, I’ve seen this worry come and go repeatedly.
Tools evolve, but what never changes is the need for developers who deeply understand both technology and the business context. Those developers have always found opportunities and will continue to do so.
The key has always been empathy—truly understanding people’s problems and distinguishing their stated wants from their actual needs. Tools are just a means to an end, and recognizing the real end goal ensures our work remains valuable, relevant, and irreplaceable.
So, rather than fearing AI, we should embrace this familiar cycle of technological advancement. By staying attentive and adaptable, we position ourselves to support businesses as they inevitably outgrow their initial AI solutions.
Ultimately, technology changes, but the human needs driving its adoption remain remarkably consistent.
Leave a Reply