Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

How I write software with LLMs

stavros.io

Post reply on HN
Print view View original

521–530 of 544 posts

Re: How I write software with LLMs

#521

Post by staticassertion »

wiseowise wrote: post #520
staticassertion wrote: post #515

Earlier quoted context omitted.

And what do you derive from that fact? The position is that coding is only one portion of the job. "But there's a coding interview" was used to rebut this position. I have pointed out that the coding interview is a fraction of the procses, once again indicating that the job involves much more than coding. So you saying "but there's a coding interview" again... who cares? Why is that relevant?

Everybody who works for salary cares. You can lament how coding is just 1% of work, it’s irrelevant what percentage is “real” coding work when you can’t pass that coding round and you’re not hired.

I have literally no clue what point you're trying to string together. I tried to refocus things to the topic at hand but you're just saying completely irrelevant things. What is your point?

Re: How I write software with LLMs

#522

Post by sarchertech »

raw_anon_1111 wrote: post #519
sarchertech wrote: post #456

Earlier quoted context omitted.

Because in a real application with real users all of the functional and non-functional requirements aren't documented anywhere but in the code.

If only a coding agent had access to your code…

You realize that coding agents aren’t AGI right? They aren’t capable of reasoning about a code changes impact on anything other than their immediate goal to anywhere near the level even a terrible human programmer is. That why we have the agentic workflow in the first place. They absolutely require guardrails.

Claude will absolutely change anything that’s not bolted to the floor. If you’ve used it on legacy software with users or if you reviewed the output you’d see this.

Re: How I write software with LLMs

#523

Post by sarchertech »

raw_anon_1111 wrote: post #511
sarchertech wrote: post #509

Earlier quoted context omitted.

Unfortunately you are not equipped to observe and test all or even most of the behavior of a non-trivial system. And if you attempt to treat every module in your system like it’s untrusted 3rd party code you’ll run into severe complexity and size limits. No one codes large systems like that because it’s not possible. There are always escape hatches and entanglements.

Actual a little company you might have heard of called Amazon does… Jeff Bezos mandated it in 2002. https://konghq.com/blog/enterprise/api-mandate AWS S3 by itself is made up of 200+ micro services

Except that they don’t. The API mandate gets violated all the time. And no one at Amazon actually treats every other team as a 3rd party.

Re: How I write software with LLMs

#524

Post by raw_anon_1111 »

sarchertech wrote: post #523
raw_anon_1111 wrote: post #511

Earlier quoted context omitted.

Actual a little company you might have heard of called Amazon does… Jeff Bezos mandated it in 2002. https://konghq.com/blog/enterprise/api-mandate AWS S3 by itself is made up of 200+ micro services

Except that they don’t. The API mandate gets violated all the time. And no one at Amazon actually treats every other team as a 3rd party.

Have you worked at Amazon?

I am not saying they treat every other team as a third party. I am saying they treat the code itself as a black box with well defined interfaces. They aren’t reaching in to another services data store to retrieve information.

Re: How I write software with LLMs

#525

Post by sarchertech »

raw_anon_1111 wrote: post #518
sarchertech wrote: post #510

Earlier quoted context omitted.

If you don’t read the code how the heck do you know anything about modularity? How do you know that Module A doesn’t import module B, run the function but then ignore it and implement the code itself? How do you even know it doesn’t import module C? Claude code regularly does all of these things. Claude code really really likes to reimplement the behavior in tests instead of actually exercising the code you told it t…

Well I know because my code is in separately deployed Lambdas that are either zip files uploaded to Lambda or Docker containers run on Lambda that only interact via APi Gateway, a lambda invoke, SNS -> SQS to Lambda, etc and my IAM roles are narrowly defined to only allow Lambda A to interact with just the Lambdas I tell it to. And if Claude tried to use an AWS service in its code that I didn’t want it to use, it wou…

This sounds like an absolute hellscape of an app architecture but you do you. It also doesn’t stop anything but the Module A imports C without you knowing about it. It doesn’t stop module A from just copy pasting the code from C and saying it’s using B.

>almost always coming into new initiatives

That says a lot about why you are so confident in this stuff.

Re: How I write software with LLMs

#526

Post by sarchertech »

raw_anon_1111 wrote: post #524
sarchertech wrote: post #523

Earlier quoted context omitted.

Except that they don’t. The API mandate gets violated all the time. And no one at Amazon actually treats every other team as a 3rd party.

Have you worked at Amazon? I am not saying they treat every other team as a third party. I am saying they treat the code itself as a black box with well defined interfaces. They aren’t reaching in to another services data store to retrieve information.

How do you know someone is ex-Amazon ? Don’t worry they’ll tell you.

I haven’t mentioned it 5 times already so you can be pretty sure I haven’t. I know too many people that have worked there to ever make that mistake.

But more importantly, have you worked there in the last decade?

Re: How I write software with LLMs

#527

Post by raw_anon_1111 »

sarchertech wrote: post #526
raw_anon_1111 wrote: post #524

Earlier quoted context omitted.

Have you worked at Amazon? I am not saying they treat every other team as a third party. I am saying they treat the code itself as a black box with well defined interfaces. They aren’t reaching in to another services data store to retrieve information.

How do you know someone is ex-Amazon ? Don’t worry they’ll tell you. I haven’t mentioned it 5 times already so you can be pretty sure I haven’t. I know too many people that have worked there to ever make that mistake. But more importantly, have you worked there in the last decade?

Then you have know idea how Amazon works. I was there from 2020-2023.

And before that I worked at 4 product companies when new to the company managers /directors/CTOs needed someone to bring in best practices and build and teach teams.

I honestly wonder what type of “big ball of mud” implementations (https://blog.codinghorror.com/the-big-ball-of-mud-and-other-...) you have had to deal with that weren’t properly componentized and covered by tests.

Re: How I write software with LLMs

#528

Post by raw_anon_1111 »

sarchertech wrote: post #525
raw_anon_1111 wrote: post #518

Earlier quoted context omitted.

Well I know because my code is in separately deployed Lambdas that are either zip files uploaded to Lambda or Docker containers run on Lambda that only interact via APi Gateway, a lambda invoke, SNS -> SQS to Lambda, etc and my IAM roles are narrowly defined to only allow Lambda A to interact with just the Lambdas I tell it to. And if Claude tried to use an AWS service in its code that I didn’t want it to use, it wou…

This sounds like an absolute hellscape of an app architecture but you do you. It also doesn’t stop anything but the Module A imports C without you knowing about it. It doesn’t stop module A from just copy pasting the code from C and saying it’s using B. >almost always coming into new initiatives That says a lot about why you are so confident in this stuff.

Yes microservice based architecture is something no modern company does…

Including the one that you were so confident doesn’t do it even though you never worked there…

Yet I don’t suffer from spooky action at a distance and a fear of changes because my testing infrastructure is weak…

Either I know what I’m doing or I’ve bullshitted my way into multiple companies into hiring me to lead architecture and/or teams from 60 person startups to the US’s second largest employer.

Did I mention that one of those companies was the company that acquired the startup I worked for before going to BigTech reached out to me to be the architect overseeing all of their acquisitions and try to integrate them based on the work I did? I didn’t accept the offer. I’ve done the “work for a PE owned company that was a getting bigger by a acquiring other companies and lead the integration thing before”

So they must have been impressed with the long term maintenance of the system to ask me back almost four years after I left

Re: How I write software with LLMs

#529

Post by kleiba »

zingar wrote: post #184
kleiba wrote: post #167

Earlier quoted context omitted.

I noticed that OpenCode requires per their own website "a modern terminal emulator" - so, no problem in Emacs? Are you running M-x term?

I have my own function that starts up a vterm in the root of the repo that I’m in. It is average for running Claude (long sessions get the scrolling through the whole history on every output character bug) but actually better at running opencode which doesn’t have this problem.

I don't have opencode installed, but I'm curious: if you start a terminal within Emacs with M-x term, how well does it do at running the opencode CLI?

Re: How I write software with LLMs

#530

Post by FeepingCreature »

raw_anon_1111 wrote: post #363
mikkupikku wrote: post #286

Earlier quoted context omitted.

Claude is surprisingly good at GUI work I've been learning, not just getting stuff working but also creating reasonably tasteful and practical designs. Asking claude in the browser to mock up a GUI and then having claude code implement it is a surprisingly powerful workflow.

I’m far away from a web developer or a web designer. But I think I intuitively understand how to put myself in the shoes of the end user when it comes to UX. I noticed that Claude is awful at understanding what makes good UX even as simple as something as if you have a one line input box and button that lets you submit the line of text, you should wire it up so a user can press return instead of pressing the button o…

yeah as it's not using its own flow you have to give it a bit of feedback. so it goes with any dev work... I think you underestimate how bad programmer uis are.
Post reply on HN