iamflimflam1

I did this recently - https://www.atomic14.com/2026/08/03/twelve-dev-boards-from-o...

Instead of trying to vibe it all - I got Claude to write deterministic scripts for creating the boards.

And did this before trying to completely vibe it: https://www.atomic14.com/2025/07/12/vibing-hardware

Let me add my elite clone - https://harmless.atomic14.com

What support do you have in your job - are you in a team of product managers/owners.

Is there a tech/team lead on the project - what’s their opinion of the progress the team is making?

It sounds like you are taking all the delivery responsibility on your own shoulders.

Is this project really top priority with your management? It sounds like it might not be.

My reading of the law was that watermarking is not required by it at all.

It’s a convenient excuse for the companies that want to add watermarking.

I really hope the improvement in natural style is real.

When I’ve tried to adjust the output style is that initially it feels better - but that’s just because the new output is so refreshing to read after the horrible Claude output.

Unfortunately, after a short while you quickly realise that it’s just as vacuous as before the style change.

This is a real problem, and it’s not just in markdown files and docs. Claude loves to write things we’ve “discovered” in comments and then later in treats the comments as gospel truth.

You have to constantly tend the garden and weed these things out.

Whenever Claude makes some error ask it where and why? And then dig out the weed.

And of course if it’s in the context, probably time for a handover doc (which will need weeding) and started fresh.

I’ve met a lot of juniors who have no idea what functional programming is.

In my experience, every company is great until things start going wrong. And when things start going wrong, no amount of culture is going to save you.

I listened to an interview with the Icelandic PM and this seems pretty much correct. She was pro EU but equally was very clear that there were other things the government could focus on instead.

Do you want us to burn a huge amount of time and effort on this. Or would you prefer us to focus on other problems?

Not much different from the UK to be honest. The constituency and first past the post system means counting is a nicely decoupled distributed system.

Each constituency declares the local winner and the only outside impact is the total count of MPs for each party.

I would prefer something more complex that does give some of the smaller parties some MPs but no one seems particularly keen to change things.

They need time to build a cursor equivalent.

Lots of enterprises don’t want to use OpenAI or Anthropic directly. They will go via AWS bedrock or google’s equivalent.

The model providers are in a difficult position. With an added danger of open weight models that are within reach of large enterprises running themselves.

This really matches up to my experience on long research projects with Claude.

It’s very hard to remove information - Claude has a habit of recording things all over the place and will happily treat things as facts even after they’ve been disproved.

What is currently true can get easily contaminated with old “facts”.

Maybe - but I would argue that the skill comes in knowing if the architecture is appropriate and knowing that LLM is not screwing it up.

I got this way with google (when it was still good) - what nonsense can I type and still get the right result - 99% of the time it would work.

Writing code with LLMs takes discipline and is difficult. You need to learn how to do it and how to get the best from it.

Software engineering is dead, long live software engineering!

That only works if the client looks at it. The current Claude fetch system does not.

Somewhere, there’s a slide deck that shows future revenue going hockey stick.

Closest thing would be the “El Farol Bar problem”

https://en.wikipedia.org/wiki/El_Farol_Bar_problem

I would assume that if you were using nitter or xcancel it’s because you did not want to support X.

You should be aware that linking to X content (whatever mechanism you use) supports it and encourages people the carry on posting there.

Just stop consuming content from X.

A lot of low level lawyer work is being replaced by AI.

My meatbag or robot scored it as likely a robot and most likely Claude.

https://meatbag.atomic14.com/

As someone who is now skewing towards the older side of the tech world (55 this year).

It’s a journey. There isn’t some point where you’ll suddenly find you are now “wise” and understand yourself.

Those moments do appear - but a few years later you’ll likely realise that it was only relevant for that time of your life and things have changed. What might have worked and served you well then may not be the right thing for now.

Getting locked into a way of thinking or behaving is a recipe for being old before your time.

Try and apply the agile principles - do a retro - is it still working, what should keep doing, what should you stop doing?

And - the best time to start doing something good is now. Doesn’t matter if you should have done it 20 years ago. Regret can be disabling.

Try and be nice to people. We don’t know what is happening in other people’s worlds.

[edit] And be kind to yourself. Self compassion is underrated. Forgive yourself. You are human. Not perfect. And you don’t need to be. Show yourself some love.

I’ve got a PhD. I suspect the problem might be people blow too much smoke up these people.

Done!

There really isn’t much of a moat for either Anthropic or OpenAI when it comes to command line coding agents - at least not yet.

I think we’re going to see some aggressive attempts at lock in quite soon. What choice do they have?

There’s a point where you have to start paying people back the money they gave you.

I’ve been in companies that have grown fast and then realised that their margins are very slim. It’s a very grim position to be in.

Great turnover - where’s the profit?

You have to really tend the garden of everything it has written.

Random off the cuff comments or one off instructions can get recorded.

And from then on they are often treated as carved in stone commandments.

It will glom onto the tiniest thing and extrapolate from it.

The sad thing about this is that getting a PhD does involve knuckling down and working hard.

I wonder what goes wrong.

Also a very trivial piece of work for an LLM to do - with a measurable end goal.

Hopefully a human did give a quick look to make sure it didn’t just delete the tests.