konschubert

Sorry, you’re right it’s bad behaviour and I’ll stop it.

But I’m not really pushing an agenda here. I use Claude a ton myself.

I think that bigger context windows help, but I feel that for AI to cross this chasm, it needs to be able to encode more abstract context knowledge. I think the holy grail here is online learning.

I think my point is that its sometimes (!) easier to use the manual chisel rather than go for the automatic chisel and then fix its mistakes.

It's the uncanny valley of AI. It's still not quite good enough yet that you can trust it blindly on a big codebase, so you still have to read and understand everything - which is often harder than just writing it up yourself.

EDIT: don’t get me wrong. I still think AI is incredibly useful for a lot of tasks! But when implementing an architecturally hairy thing, I find it less stressful and equally quick to jump down to the editor level and use AI just for code completion.

It's the uncanny valley of AI. It's still not quite good enough yet that you can trust it blindly on a big codebase, so you still have to read and understand everything - which is often harder than just writing it up yourself.

I think this misses something.

AI is going to struggle at building a consistent internal model of the domain into the software unless you’re able to give a structured explanation of the domain.

If you’re just giving it a set of inputs and expected outputs, it’s not going to generalise well and fail at out of sample input, unless the AI already understands the domain from its training set.

Being able to give a structured explanation of a domain (and being able to judge if the internal model of the software makes sense) is not the same as having experience in a domain.

Lots of ppl with domain experience can tell a right output from a false one, but can’t tell you why.

This isn’t a response to you, but just so it has been said: 3 to 8-year-old kids have no business being on YouTube alone. AI or not.

You are contradicting yourself. If you're hoarding the data for yourself you're not going to develop something useful. Sharing the data means that it will be integrated into the big LLMs, which will be useful "for their own country".

More likely there would have been one less job.

There isn't a "lump of labour" that gets distributed in the economy.

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

This is called the "lump of labour" fallacy:

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

My point is that this is clearly out of the lab.

I think CATL bringing sodium-ion to industrial scale should count as "on the shelf".

https://www.pv-magazine.com/2026/04/28/catl-secures-worlds-l...

I don't know what to say. I may not like it, you may not think it's actually intelligent, you may not think it's going to change the world - but how can you not see that this is revolutionary?

“Insurance”

Check out Technology Connections. This is way, way, way more in depth than anything one can find on TV.

Maybe they can:

https://news.ycombinator.com/item?id=48071262

I’m not saying that you’re wrong.

But it’s worth noting that any good technology starts off being called a toy and with most people not being able to imagine its usefulness.

I hope that we will soon have ways to change the tone of AI writing, I hate that all news articles now have that same AI voice.

> that teaches you how to manipulate people

Well, it's called "... and influence people", so I see where you were coming from in your assumption.

> do your hobby with other people, frequently

I think this advice can be refined.

The goal is to meet people with shared interests, regularly.

You may like to work out, but unless you enjoy talking and thinking about working out, the gym isn't the right place!

You may like to hike, but you may not enjoy talking and thinking about hiking. You may still find friends in the hiking group, because as you spend time together you discover shared interests. But you may also discover that you have very little in common with the people who like hiking.

That's why it's so easy to make friends in University: You spend a lot of time together AND you have a shared interest: The major you all chose!

On the other hand, if you're into ... trains... you need to find places where people go that like to talk and think about trains! That's not always easy!

What isn't cheap, though, is maintaining the plumbing of the communal system, especially outside cities.

Friends in south Sweden and they got a hole drilled in the front yard like it’s the most normal thing. Is it there?

I built an optimisation for charging my car. It’s very smart, looks multiple days ahead in weather and prices and predicts my driving habits.

I tried to make it a product, but I didn’t find much interest. Maybe I’m just bad at marketing.

With the help of Claude, it cost me years of experience and few weekends to build. So even if nobody other than me ever profits from it, it was still worth it!

I’m a soft developer by trade and CAD is just something I do in order to be able to make the hardware I want to make. I’m not good at it and it takes me forever.

This is very exciting.

Next step, I need the same for PCB design, with good understanding of EMI issues.

Okay, I will read it and then probably need to google most of these :D

Is the internal data model of fusion structured enough to be understood with a text-based LLM? Or do you need to basically screenshot the render to understand what is happening?

Would a more CAD-as-code based approach to CAD design be more suitable?

Just like, LLMs have an easier time to build a presentation with latex than with powerpoint...

These standalone batteries are typically privately financed.

The noon-to-evening spread on the German day ahead market today in the 1st of May is 700 Euro per MWh.

Yea. With a huge 100 kWh battery and a removable range extender for those extra-long trips :) Plus that battery (and range extender) can also provide power and heating when parked.

Hi,

I think that's a fair point, I underestimated the power needs.

For a 1GW data center, the economical case is probably to build about 2 GWp of solar, plus a 5GWh battery or so.

That is not unrealistic:

https://masdar.ae/en/news/newsroom/masdar-and-consortium-par...

https://www.pv-magazine.com/2026/04/28/catl-secures-worlds-l...

On a sunny day, that will let you to power the data center fully from solar for most of the day, and partially through the night.

Obviously you need ~1GW of gas turbine capacity as well, for the non-so-sunny-days, so the economic case for solar here is dependent on the fuel price.