Do you see departments at your company building small apps? | Hacker News Reader

Do you see departments at your company building small apps?

How often are you seeing non-technical people at your company building their own software to solve everyday problems? I saw this at my last company. They even created a department within tech just to handle it. Is this happening everywhere? What do you think?

Refs: https://www.ycombinator.com/rfs#a-cloud-for-small-software https://www.shpyrd.io/small-software-manifesto

View on HN

It has been happening for my entire 30+ year career. This is not a new trend. LLMs are the newest flavor of it, but low-code/no-code tools have been around for decades. As well as clever people who just figure out how to create apps.

They have a really standard lifecycle: They get built. They work for a small department. If it stops there, it just keeps running until the person who built it moves on, then it gets trashed and replaced with something else. Or, if it grows beyond the department, that person's life starts to get taken up in maintaining it, it fails to scale, and IT takes it over (or replaces it).

The question is not does it happen. It happens so often that there is even a term for it: "Shadow IT". The question is not even whether or not you can launch a product to stabilize the lifecycle. The real question is - should you? The ability to create an app doesn't grant the knowledge of how to make it secure and fit within regulatory bounds for your industry. It is quite possible (and happens often) for someone to make an app for their department that works great and everyone loves, but exposes the company to real regulatory and legal risks.

That is the real reason IT shops have someone to watch over such apps. As you are clearly asking this question in the spirit of self-promotion, you really need to show that you understand the real problems and risks with departmental apps. "Do people build them?" makes you sound like you are completely new to how corporate IT works, which instills no confidence that you have built a viable product.

"They have a really standard lifecycle: They get built. They work for a small department. If it stops there, it just keeps running until the person who built it moves on, then it gets trashed and replaced with something else. Or, if it grows beyond the department, that person's life starts to get taken up in maintaining it, it fails to scale, and IT takes it over (or replaces it)." This is how it was happening, its very different now and that's the reason why I'm asking this.

Since you think I'm new to this, lemme get you some background. I'm an experienced founder, built companies with 150+ employees and raised capital from big names from industry (https://www.bloomberg.com/news/articles/2020-10-20/goldman-b...), figured on TechCrunch also. So you need to stop assume people are new to things just because they formatted their post to be more beginner friendly ;).

Things changed a lot. With AI, you can plug-in seamless between the deploy and the actual hosting, you can automate putting it online in the local private subnet, you can even change code in the middle of deploy using AI and applying local best practices, change the database and any underlying implementation and stuff like that. This isn't happening your entire 30+ year career, sorry Dave.

PS: The company I was referring too is one of which I founded, and before moving to the board it had like 80+ apps built from departments. And I know other founders with exactly the same behavior.

Thank you for your time

Reply on news.ycombinator.com