Live data from Hacker News

Coding agents have replaced every framework I used

blog.alaindichiappari.dev

281–290 of 611 posts

Re: Coding agents have replaced every framework I used

#281

Earlier quoted context omitted.

Now run that loop 1000 times. What does the code /system look like. It is going to be more like evolution (fit to environment) than engineering (fit to purpose). It will be fascinating to watch nonetheless.

Sure, if all you ask it to do is fix bugs. You can also ask it to work on code health things like better organization, better testing, finding interesting invariants and enforcing them, and so on. It's up to you what you want to prioritize.

Your assuming that scrum/agile/management won't take this over?

What stakeholder is prioritizing any of those things and paying for it out of their budget?

Code improvement projects are the White Whale of software engineering - obsessed over but rarely from a business point of view worth it.

Re: Coding agents have replaced every framework I used

#282

A significant number of developers and businesses are going to have an absolutely brutal rude awakening in the not too distant future. You can build things this way, and they may work for a time, but you don't know what you don't know (and experience teaches you that you only find most stuff by building/struggling; not sipping a soda while the AI blurts out potentially secure/stable code). The hubris around AI is goi…

Have you considered that betting against the models and ecosystem improving might be a bad bet, and you might be the one who is in for a rude awakening?

I agree. We've been assured by these skeptics that models are stochastic parrots, that progress in developing them was stalling, and that skills parity with senior developers was impossible - as well as having to listen to a type of self-indulgent daydreaming relish about the eventual catastrophes companies adopting them would face. And perhaps eventually these skeptics will turn out to be right. Who knows at this stage. But at this stage, what we're seeing is just the opposite: significant progress in model development last year, patterns for use being explored by almost every development team without widespread calamity and the first well-functioning automated workflows appearing for replacing entire teams. At this stage, I'd bet on the skeptics being the camp to eventually be forced to make the hard adjustments.

Re: Coding agents have replaced every framework I used

#283
post #255

A framework also gives you someone's expertise in a domain so you don't have to develop that expertise yourself and focus on all the other stuff...

...and importantly, neither does the LLM; frameworks are incredibly useful even if you are using generative AI.

Re: Coding agents have replaced every framework I used

#284

A significant number of developers and businesses are going to have an absolutely brutal rude awakening in the not too distant future. You can build things this way, and they may work for a time, but you don't know what you don't know (and experience teaches you that you only find most stuff by building/struggling; not sipping a soda while the AI blurts out potentially secure/stable code). The hubris around AI is goi…

The future is already here. Been working a few years at a subsidiary of a large corporation where the entire hierarchy of companies is pushing AI hard, at different levels of complexity, from office work up through software development. Regular company meetings across companies and divisions to discuss methods and progress. Overall not a bad strategy and it's paying dividends. A experiment was tried on a large and ve…

These are the blog posts we need.

This is the siren song of llm. "Look how much progress we made"

Effort increases as time to completion decreases. The last 10% of the project takes 90% of the effort as you try to finish up, deploy,integrate and find the gaps.

Llms are woefully incapable of that as that knowledge doesn't exist in a markdown file. It's in people's heads and you have to pry it out with a crowbar or as happens to so many projects, they get released and no one uses it.

See Google et Al. "We failed to find market fit on the 15th iteration of our chat app, we'll do better next time"

Re: Coding agents have replaced every framework I used

#285

Earlier quoted context omitted.

>software engineers today are 100x more productive Somebody needs to explain to my lying eyes where these 100xers are hiding. They seem to live in comments on the internet, but I'm not seeing the teams around me increase their output by two orders of magnitude.

They are the people who have the design sense of someone like Rob Pike but lack his coding skill. These people are now 100x more capable than they were previously.

No they’re not.

Re: Coding agents have replaced every framework I used

#286
post #157

Earlier quoted context omitted.

>A significant number of developers and businesses are going to have an absolutely brutal rude awakening in the not too distant future. Correct. Those who wave away coding agents and refuse to engrain them into their workflows are going to be left behind in the dust.

> Correct. Those who wave away AI and refuse to engrain it into their workflows are going to be left behind in the dust. Similar to those who waved away crypto and are now left behind in the dust, yes?

Might not be the best counter example since everyone who has bought BTC before Jan 2024 is now in massive profit.

Re: Coding agents have replaced every framework I used

#287

Earlier quoted context omitted.

I think Bitcoin and major cryptos outperformed a lot of assets over the last decade, so you could say it left some people behind in the dust, yes

Like being ratioed with a 50% price crash?

You mean just like META, NFLX, AMZN, TSLA, NVDA, CSCO, MSFT, GE, BAC ?

Re: Coding agents have replaced every framework I used

#288

A significant number of developers and businesses are going to have an absolutely brutal rude awakening in the not too distant future. You can build things this way, and they may work for a time, but you don't know what you don't know (and experience teaches you that you only find most stuff by building/struggling; not sipping a soda while the AI blurts out potentially secure/stable code). The hubris around AI is goi…

The aspect of "potentially secure/stable code" is very interesting to me. There's an enormous amount of code that aren't secure or stable already (I'd argue virtually all of the code in existence).

This has already been a problem. There's no real ramifications for it. Even for something like Cloudflare stopping a significant amount of Internet traffic for any amount of time is not (as far as I know) investigated in an independent way. There's nobody that is potentially facing charges. However, with other civil engineering endeavors, there absolutely is. Regular checks, government agencies to audit systems, penalties for causing harm, etc. are expected in those areas.

LLM-generated code is the continuation of the bastardization of software "engineering." Now the situation is not only that nobody is accountable, but a black box cluster of computers is not even reasonably accountable. If someone makes a tragic mistake today, it can be understood who caused it. If "Cloudflare2" comes about which is all (or significantly) generated, whoever is in charge can just throw their hands up and say "hey, I don't know why it did this, and the people that made the system that made this mistake don't know why it did this." It has been and will continue to be very concerning.

Post reply on HN