Even if only law enforcement can use it, having that feature is highly regulated.

[edit] I see this is from years ago. I should read the articles first. :)

It helps formalise your plan, then creates some code, you review, talk about what you believe to be wrong or asking it why it took that approach, or even telling it to take the approach that you want.

The end result a world of difference, and I feel I have a better grasp of what is going on in the whole application.

The technology is accelerating. Hard projects from early last year are now trivial for me. Even AI related tools we are using internally are being made redundant from open source models and new frameworks (eg. OpenClaw).

It feels like we are in the AI version of "Don't look up". Everyone is on borrowed time, you should be looking at how to position yourself in an AI world before everyone realises.

The two main takeaways. Create a CLAUDE.md file that defines everything about the project. Have Claude feed back into the file when it makes mistakes and how to fix them.

Now it creates well structured code and production level applications. I still double check everything of course, but the level of errors is much lower.

An example application it created from a CLAUDE.md I wrote. The application reads multiple PDF's, finds the key stakeholders and related data, then generates a network graph across those files and renders it in an explorable graph in Godot.

That took 3 hours to make, test. It also supports OpenAI (lmstudio), Claude and Ollama for its LLM callouts.

What issue I can see happening is the duplication of assets in work. Instead of finding an asset someone built, people have been creating their own.

Instead have Claude know when to offload work to local models and what model is best suited for the job. It will shape the prompt for the model. Then have Claude review the results. Massive reduction in costs.

btw, at least on Macbooks you can run good models with just M1 32GB of memory.

Although I'm starting to like LMStudio more, as it has more features that Ollama is missing.

https://lmstudio.ai

You can then get Claude to create the MCP server to talk to either. Then a CLAUDE.md that tells it to read the models you have downloaded, determine their use and when to offload. Claude will make all that for you as well.

The big powerful models think about tasks, then offload some stuff to a drastically cheaper cloud model or the model running on your hardware.

Building a program language is like poetry. Everyone does it at some point, just most of us know never to share it.

The majority of code work is maintaining someone else's code. That's the reason it is "nicer".

There is also the matter of performance and reducing redundancy.

Two recent pulls I saw where it was AI generated did neither. Both attempted to recreate from scratch rather than using industry tested modules. One was using csv instead of polars for the intensive work.

So while they worked, they became an unmaintainable mess.

If you haven't reviewed and signed off then you have to assume that the stuff is garbage.

This is the crux of using AI to create anything and it has been a core rule of development for many years that you don't use wizards unless you understand what they are doing.

I'd say for what I'm trying to do - which is upgrade a very old version of PHP to something that is supported, this is completely acceptable. These are basically acting as smoke tests.

You need to be a bit careful here. A test that runs your function and then asserts something useless like 'typeof response == object' will also meet those code coverage numbers.

In reality, modern LLMs write tests that are more meaningful than that, but it's still worth testing the assumption and thinking up your own edge cases.

There is a documentary called "Everything under control". In it they explained why this happened.

Basically they were scared that the public were going to buy out the masks that were needed by medical staff.

> Ivermectin

Same documentary, this was started by Musk. It does nothing and is dangerous.

One question, have you ever considered the opposite of what you're saying to be true, or looked for the evidence for that? Saying because I've heard of and looked at both opinions you've expressed in your comment and heard and seen evidence for them to be true. I also did the opposite. And looking at the opposite seemed more true objectively and with the emotions and popular biases like authority bias and other harmful ones removed.

"If you can't see anything wrong with the side you agree with, and you can't see anything right with the side you disagree with, you've been manipulated." Very wise quote.

Yea sorry I was recalling from memory.

> One question, have you ever considered the opposite of what you're saying to be true, or looked for the evidence for that?

I look at what the scientists who have spent years in that field are saying and is peer reviewed.

You go searching on the internet for the answer you expect, you will find it.

> authority bias

Authority bias is where an expert in field X claims to have knowledge in field Y when they don't.

| Authority Bias: The cognitive tendency to overvalue the opinion of an authority figure or expert

lets not make up definitions for your own convenience.

Authority bias is the cognitive bias where we give disproportionate weight to the opinion of someone perceived as an authority figure, even outside their domain of expertise.

What a lot of people outside the scientific field fail to realise that when a claim is made by an expert in their field, it is peer reviewed and challenged or accepted.

The whole point of science is that you can challenge wrong ideas and change your perspective.

Like I said though. If you are just going to the internet to find something that aligns with what you believe, that isn't peer review.

Did you not share this above? This contradicts to me your definition here, which is just copying what I shared and stating it as thought its now your opinion. Comes off like gaslighting by the way.