Hacker Newsnew | past | comments | ask | show | jobs | submit | imatworkyo's commentslogin

This is awesome, good luck with this tool. Can it do isometric and top down views?


working on adding those


how often are we truly writing actual novel programs that are complex in a way AI does not excel at?

There are many types of complex, and many times complex for a human coder, are trivial for AI and its skillset.


Depends on the field of development you do.

CRUD backend app for a business in a common sector? It's mostly just connecting stuff together (though I would argue that an experienced dev with a good stack takes less time to write it as is than painstakingly explaining it to an LLM in an inexact human language).

Some R&D stuff, or even debugging any kind of code? It's almost useless, as it would require deep reasoning, where these models absolutely break down.


Have you tried debugging using the new "reasoning" models yet?

I have been extremely impressed with o1, o3, o4-mini and Gemini 2.5 as debugging aids. The combination of long context input and their chain-of-thought means they can frequently help me figure out bugs that span several different layers of code.

I wrote about an early experiment with that here: https://simonwillison.net/2024/Sep/25/o1-preview-llm/

Here's a Gemini 2.5 Pro transcript from this afternoon where I'm trying to figure out a very tricky bug: https://gist.github.com/simonw/4e208ab9edb5e6a814d3d23d7570d...


In my experience they're not great with mathy code for example. I had a function that did subdivision of certain splines and had some of the coefficients wrong. I pasted my function into these reasoning models and asked "does this look right?" and they all had a whole bunch of math formulas in their reasoning and said "this is correct" (which it wasn't).


Wait I’ve found it very good at debugging. It iteratively states a hypothesis, tries things, and reacts from what it sees.

It thinks of things that I don’t think of right away. It tries weird approaches that are frequently wrong but almost always yield some information and are sometimes spot on.

And sometimes there’s some annoying thing that having Claude bang its head against for $1.25 in API calls is slower than I would be but I can spend my time and emotional bandwidth elsewhere.


I agree with this. I do mostly DevOps stuff for work and it’s great at telling me about errors with different applications/build processes. Just today I used it to help me scrape data from some webpages and it worked very well.

But when I try to do more complicated math it falls short. I do have to say that Gemini Pro 2.5 is starting to get better in this area though.


is this really what you are using AI for ?


Google recently said transformers wouldn't work ...


looking at this response, https://news.ycombinator.com/item?id=19307693

it seems to use your own stem cells, and is not used because of its risk factor


there are 168,000 gas stations in America. Any given day maybe 100 have a skimmer working...seems like an moderate inconvenience for something that will literally probably never happen to you.

And if it does... just call your bank and get a new card and get the transactions reversed in a 30min phone call.


You're assuming a random distribution of skimmers at pumps across the nation.

Such criminal activity tends to occur in concentrated pockets.

Gas stations around Joshua Tree, CA had been rumored to be skimming people's cards, and some now carry large official stickers warning users of their cards likely getting frozen for 72 hours should they be used at the pump, confirming the rumors.

I imagine the active number is probably more in the thousands nationally. There are a lot of lower income, high-crime areas where pumps are neglected. The Venn diagram of gas stations lacking squeegees and those compromised I suspect has a lot of overlap.


> How did that happen?

Outside of the false equivalence - as posted to great fanfare yesterday :

https://i.redd.it/1j5ek052qit01.jpg


Anyone know what that looks like if showing person-years instead of years? Curious the effect that integrating over population would have. Could imagine integrating over some inflation or return measures too.


Definitely cool service - do these numbers support MMS?


Not at the moment no. I might look into it before launch but to be honest nowadays there is very little business case for MMS so I’m not sure it’s worth the investment.

Just out of curiosity, what do you use MMS for?


have done it plenty of times. Just call the driver and ask if they mind having a package instead of a passanger (no one will care).

And yes make sure someone is ready to receive the items, then you are golden.


that might be comparing two different things. In a single parent household (which I'm sure is heavily swayed towards single mothers) kids may be out playing with other kids.

In families with older fathers, [assuming] there is a higher percentage of dual parenting ... which could mean more attention at home = less time playing sports with the neighborhood kids


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: