Robots will eventually be capable of taking all of our jobs, but let’s just put that aside for a minute to talk about legitimate vibe-coding risks.
A lot of people are complaining about vibe coding. And making fun of it. And trashing it. And grabbing torches and pitchforks to march to their nearest data center.
Well, we might not be at the pitchfork thing yet, but we’re close. There are a lot of vibe-coding risks, but also a lot of overreactions.
The Reason For The Overreactions
My powers of observation say that people (especially developers/engineers) hate the whole vibe-coding thing for these three reasons:
- They’re scared it will take their jobs
- They think they’re superior to others, and this makes them feel threatened
- They’re sick of goofballs saying “My vibe-coded app will now kill [insert profitable enterprise tool here]!!!!!!”
The first two are psychological things, and I don’t have the energy to get into that today. But let’s look at the third part.
Can A Vibe-Coded App Kill An Enterprise Tool?
Anything is possible, but a hacked together app isn’t going to knock out a multi-million dollar product. If nothing else, it doesn’t come with the necessary infrastructure to do that, and it would have to convince people to change.
When Can Vibe-Coded Apps Solve Business Problems?
The main arguments I hear against an app are that they’re not super safe, prepared for every contingency, or monitored 24/7.
Those are all valid, but they overlook a challenge I’ve faced my entire professional career. I’ve had so many cases of needing a very niche and specific tool, but I don’t have the budget (and/or it doesn’t exist).
My Real-Life Example
In my current job, I have needed a specific (and simple) app for a while. There’s no budget to get one, the team doesn’t want to adapt to a new one, and I’m really the only one who suffers under the current setup. So I’ve been stuck in a frustrating Google Sheets process.
It wastes a ton of my time, and it’s not very accurate.
So, I spent a weekend with Claude Code, and I built a very simple app that solves the problems on my end. To avoid adoption by coworkers, I used Google auth to login, so there wasn’t any friction for my teammates to join.
This is a tool that I was capable of building myself, but I didn’t have the time when I had to manually code every single line, and troubleshoot every single error.
Why It’s Good Enough
This one fits my use case for many reasons.
I would normally worry about safety, but not here. Only a handful of people need this tool, and it doesn’t contain any sensitive info. Risks are low.
The speed of development was supersonic. With Claude’s help, I built it start to finish in the free time of my weekend.
Times Are Changing
In the old days, I wouldn’t have been able to have this tool. I would’ve just suffered at work.
So, I get where people are coming from when they complain about this whole vibe-coding thing. It’s a mixture of seeing weaknesses and feeling threatened.
In many cases (like my example), these won’t replace any one’s job. They would’ve just been ignored in the past, but they now have a simple tool where good enough is good enough
My whole point is to stop worrying and just focus on making your own products better.
Oh yeah, and if you sign up for a “[enterprise tool] killer!!!!”, be skeptical about the quality.
