Reminds me of my current customer.
We (another freelancer and me) built an application that replaced an Excel sheet, which was the foundation of the business until then. So the usual so far.
We have a policy that our customers are responsible for all their business-related input, but we make the decisions about the technical implementation.
Every technical decision that the customer wants to make basically costs extra.
In this case we built a rather simple multi-tenancy B2B app using Laravel, with one database per tenant.
They planned to start with a single customer/tenant, scaling up to maybe a few dozen within the next years, with less than a hundred concurrent users over the first five years.
There were some processes with a little load, but they were few, running less that a minute each and already built up to run asynchronous.
We planned a single Hetzner instance and to scale up as soon as we would see it reaching its limits.
So less than 100 €/month.
The customer told us that they have a cooperation with their local hosting provider (with "special conditions!") and that they wanted to use them instead.
My colleague did all the setup, because he is more experienced in that, but instead of our usual five-minute-setup in Forge (one of the advantages of the Laravel ecosystem), it took several weeks with the hosting provider, where my colleague had to invest almost full time just for the deployment.
The hosting provider "consulted" out customer to invest in a more complex setup with a load balancer in front, to be able to scale right away.
They also took very long for each step, like providing IP addresses or to handle the SSL certificates.
We are very proud of our very fast development process and having to work with that hosting provider cost us about one third of our first development phase for the initial product.
It's been around two years since then.
While the software still works as intended, the customer could not grow as expected. They are still running with only one single tenant (basically themselves) and the system barely had to handle more than two concurrent users.
The customer recently accidentally mentioned that they pay almost 1000€/month for the hosting alone.
But it scales!
I'm usually first in line when talking shit about the German government, but here I am absolutely for this.
I was really positively surprised when I had my apprenticeship at a publishing company and we had a routine to bring physical backups to the cellar of a post office every morning.
The company wasn't that up-to-date with most things, but here they were forced to a proper procedure which totally makes sense.
They even had proper desaster recovery strategies that included being back online within less than 2 hours hours even after a 100% loss of all hardware.
They had internal jokes that you could have nuked their building and as long as one IT guy survived because he was in the home office, he could at least bring up the software within a day.
Ha, I've written a lot of these Makefiles and the "make dev" command even became a personal standard that I added to each project. I don't know if I read about that, or if it just developed into that because it just makes sense.
In the last few years, these commands very often started a docker container, though.
I do tend to work on Windows with WSL and I most of my colleagues use macOS or Linux, so that's definitely one of the reasons why docker is just easier there.
Absolutely!
I have used Copilot for a few weeks and then stopped when I worked on a machine that didn't have Copilot installed and I immediately struggled with even basic syntax.
Now I often use LLMs as advanced rubber ducks. By describing my problems, the solution often comes to my mind on its own and sometimes the responses I get are enough for me to continue on my own.
In my opinion, letting LLMs directly code can be really harmful for the software developers, because they forget to think for themselves.
Maybe I'm wrong and I am just slow to accept the new reality, but I try to keep writing most of my code on my own and improve my coding skills more than my prompting skills (while still using these tools, of course).
For me, LLMs are like a grumpy and cynical old senior dev who is forced to talk in a very positive manner and who has fun trickling in some completely random bullshit between his actual helpful advice.
I guess it totally depends on the specific use cases.
Just yesterday, I asked Gemini 2.5 for how to test a Laravel artisan command that internally calls another artisan command using the Pest testing framework.
Gemini showed me a method that looked perfect.
But when I tried to execute it, I got an error that the method doesn't exist.
So I asked Gemini about this, and it told me: "Ah, apologies! That's a classic sign that you might be using a version of Laravel older than 5.7."
So I told it that I am actually using the latest release (12.3), to which it responded with: "You're right to question things when you get an error! However, let's clarify the Laravel version first.
As of today, March 27, 2025, the latest stable version of Laravel is Laravel 11, which was released in March 2024. Laravel 12 has not been released yet (it's expected roughly in Q1 2026 based on the usual schedule).
Could you please double-check the exact Laravel version you are using?"
So it did not believe me and I had to convince it first that I was using a real version. This went on for a while, with Gemini not only hallucinating stuff, but also being very persistent and difficult to convince of anything else.
Well, in the end it was still certain that this method should exist, even though it could not provide any evidence for it and my searching through the internet and the Git history of the related packages did also not provide any results.
So I gave up and tried it with Claude 3.7 which could also not provide any working solution.
In the end, I found an entirely different solution for my problem, but that wasn't based on anything the AIs told me, but just my own thinking and talking to other software developers.
I would not go that far to call these AIs useless. In software development they can help with simple stuff and boilerplate code, and I found them a lot more helpful in creative work. This is basically the opposite from what I would have expected 5 years ago ^^
But for any important tasks, these LLMs are still far too unreliable.
They often feel like they have a lot of knowledge, but no wisdom.
They don't know how to apply their knowledge ideally, and they often basically brute-force it with a mix of strange creativity and statistical models that are apparently based on a vast amount of internet content that has big parts of troll content and satire.
I have met several of these leaders. Problem is, that very few of them are also very successful.
Most of these "leaders" that I met, based a lot of their success from basically being able to stack their bs very high and bailing out before that stack fell over.
Interestingly, the one guy that I am the most sure of being a good guy and leader is also one of the most successful, having retired with several hundred millions in the bank.
But the most successful guy is the one with the worst methods, who is very close to become a billionaire (he might actually already be one, haven't checked on him for a few months).
This is the first time that I saw this blog and I've now read a bunch of those posts.
There is so much gold there, it's incredible.
So much is going wrong in all those companies, that it is just frustrating to be working in those.
A colleague of mine and me got the opportunity a while back to basically work as wanted to in one side project.
So we just worked truly agile, without any of the "Scrum" bs around it.
Everyone involved was blown away how successful it was in terms of implementation.
Two companies tried it before in the span of a year, and they didn't get it working at all.
We worked just some evenings and weekends outside of our full-time job over the span of six weeks, and we got 90% done, with everything essential included to be able to start working with it.
We then implemented everything else the customer wanted within the next weeks by following the same principles.
There were no problems at all (apart from the customer basically almost running out of money because they spent most of on the previous companies which couldn't deliver).
At my day job, the managers usually argue against anything we want to do, question every decision from our side, want to have regular big meetings with 16+ people to talk about everything at length until no one wants to do it anymore, and then micro-manage everyone to death.
We tried to at least keep these guys out of the daily by pointing to the Scrum guide where it says that only people working on tickets should participate, so one of these guys just created a bs task for himself, then talked every day about it for a lengthy amount of time without providing any real value at all.
I never got those silly shows before, which play in offices where everyone just acts silly and they never really work.
But now I know why they are so popular. They just provide comic relief for people who really have to work in those environments and it is really as bad as in the television shows, if not worse.
I've been searching through the comments for a comment about this item in the list.
I am less than 2 meters tall, but I can carry plates wider than 2 meters between my outstreched hands.
Does that now mean that I am an ape? Actually, that would explain a lot...
We have a policy that our customers are responsible for all their business-related input, but we make the decisions about the technical implementation. Every technical decision that the customer wants to make basically costs extra.
In this case we built a rather simple multi-tenancy B2B app using Laravel, with one database per tenant. They planned to start with a single customer/tenant, scaling up to maybe a few dozen within the next years, with less than a hundred concurrent users over the first five years. There were some processes with a little load, but they were few, running less that a minute each and already built up to run asynchronous.
We planned a single Hetzner instance and to scale up as soon as we would see it reaching its limits. So less than 100 €/month.
The customer told us that they have a cooperation with their local hosting provider (with "special conditions!") and that they wanted to use them instead.
My colleague did all the setup, because he is more experienced in that, but instead of our usual five-minute-setup in Forge (one of the advantages of the Laravel ecosystem), it took several weeks with the hosting provider, where my colleague had to invest almost full time just for the deployment. The hosting provider "consulted" out customer to invest in a more complex setup with a load balancer in front, to be able to scale right away. They also took very long for each step, like providing IP addresses or to handle the SSL certificates.
We are very proud of our very fast development process and having to work with that hosting provider cost us about one third of our first development phase for the initial product.
It's been around two years since then. While the software still works as intended, the customer could not grow as expected. They are still running with only one single tenant (basically themselves) and the system barely had to handle more than two concurrent users. The customer recently accidentally mentioned that they pay almost 1000€/month for the hosting alone. But it scales!