Here’s the raw truth: I was a developer for the first 10 years of my professional life. I’ve always loved to implement new things and extend them – that was pure passion, but it became clear that with off-shoring I was replaceable. When I gave up on my developer career and moved into business areas (project management, product management), it was difficult to get used to the fact that I no longer controlled the technology, and this often frustrated me.
Fast forward to today – Claude Code rekindled my passion for software development. Well, I’m not actually writing code, I’m just asking Claude Code to write it for me. What I learned in that process is that I was never really passionate about programming, I was passionate about the outcomes. But not just that: the reason I was frustrated in my previous roles is because the outcome was always several months away. This was not good for my ADHD brain – I’m driven by the things that happen now.
Enter Claude Code – my first question: where the heck have you been all my life? Seriously, this passion of mine for creating things in code, I have to admit, is maybe taking on a new form that I could compare to addiction. Here’s what makes it so addictive: I can basically do anything in a short amount of time that would have taken me days or weeks to implement. But not just that, I can do things now that I simply didn’t know how to implement before.
I spend my weekends working on my personal projects and I’m sometimes vibe coding for 8 hours straight. That can’t be healthy, can it? But more to the point: I vibe code something and it’s done in 5 minutes, and it opens up new possibilities that I then vibe code again.
Case in point: I was working on my personal CRM app, which I have to proudly say is getting really good. I can do campaigns, I have an integration with my calendar and email, I have APIs which I use externally – for example, I have my own “schedule an appointment with me” app, etc. There’s a lot of stuff in there already. Yesterday, I decided I wanted note-taking capabilities similar to Obsidian. A good 18 hours later – I have it. I’ve basically built something like a clone of Obsidian. But that’s not enough: I also wanted to make it work as a progressive web app with offline capabilities. It took me 2 hours to do that. And this now opens up new capabilities, like using the camera on my phone – so my next idea is to be able to scan business cards, OCR them, and add that contact. The ideas simply don’t stop.
But stepping back and looking at the bigger picture, I see two dramatic transitions.
- Software development is moving in-house again, but this time it’s the software engineers that become more like product owners, and instead of using off-shore development centers, they use Claude Code (or similar). The reason you want a software engineer at the core of this is because you still need people who understand the technology, and especially the architecture.
- Large (and medium-sized) companies will do the math: how much does a “seat” cost on a SaaS platform vs. building it ourselves? I think that’s the biggest insight for me: I can write a complete CRM app from scratch. But not just that – I can dictate precisely which features I want and which I don’t. Now, I’m not saying this is the end of SaaS, because these platforms are still good value for companies that are not in a position for self-determination. But if you have the determination and the business domain knowledge, what would stop you from implementing this on your own? And it wasn’t that long ago that large companies built their own software – I think we’re seeing a revival of this.
What do you think?
Comments
Post a Comment