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

AI demands more engineering discipline. Not less

charitydotwtf.substack.com

Post reply on HN
Print view View original

211–220 of 234 posts

Re: AI demands more engineering discipline. Not less

#211

Post by trjordan »

gavinh wrote: post #101
trjordan wrote: post #27

> Those are not code problems. They are evaluation problems. > Code becomes precious when it is the only place knowledge lives. Reading AI code all day is _agonizing_. Just, a horrible way to live, and it melts people's brains at the moment you need them to be the most capable. Manual programming has this really productive and gratifying feedback loop, where you read the code, write the code, and fix it until it comp…

I agree that reading AI code all day is agonizing. We're relying on code review to develop parts of our mental model of the system that were previously developed through coding. We're having more difficulty comprehending and recall details of the system. This is probably unsurprising; people recall information better that they "generated" than information they read. I am applying some lessons from pedagogy to extend…

Would love to chat -- ping me tr at tern dot sh

Re: AI demands more engineering discipline. Not less

#212

Post by ChrisLTD »

win311fwg wrote: post #182
ChrisLTD wrote: post #151

Earlier quoted context omitted.

How can the code no longer matter? It literally is the logic (not to mention performance, and reliability) of the software.

You might say in the same way that machine code stopped mattering when programming languages gained in popularity. Almost nobody will ever review machine code. I anticipate 90% of all programmers today wouldn't even know how. The move again is towards a higher level of abstraction; this time validation. Instead of describing how the program is to function, you define the properties of the system and let the fancy com…

I can be 99.99999999% certain when I write an if statement like "if (x > 1) do y" that the compiler will turn that into the equivalent machine code. So, yes, unless I hit some crazy performance bottleneck, I'm not concerned about reviewing the machine code.

However, LLM outputs change with slight re-wording of prompts and with each new model release. I could hand write a test that says if x > 1 make sure y happens, but then what productivity was gained?

Re: AI demands more engineering discipline. Not less

#213

Post by win311fwg »

ChrisLTD wrote: post #212
win311fwg wrote: post #182

Earlier quoted context omitted.

You might say in the same way that machine code stopped mattering when programming languages gained in popularity. Almost nobody will ever review machine code. I anticipate 90% of all programmers today wouldn't even know how. The move again is towards a higher level of abstraction; this time validation. Instead of describing how the program is to function, you define the properties of the system and let the fancy com…

I can be 99.99999999% certain when I write an if statement like "if (x > 1) do y" that the compiler will turn that into the equivalent machine code. So, yes, unless I hit some crazy performance bottleneck, I'm not concerned about reviewing the machine code. However, LLM outputs change with slight re-wording of prompts and with each new model release. I could hand write a test that says if x > 1 make sure y happens, b…

Just like you can be 99.99999999% certain that your properties will eventually be satisfied when the fancy compiler turns it into equivalent code.

If your program is truly nothing more than that "if" statement then there is unlikely to be any productivity gain, just as there is no real productivity gain using a programming language over flipping toggle switches for something so simple. Programming languages would have never been invented if that bit of logic summed up the entirety of computer science. In the real world, the calculus starts to change when you are trying to solve bigger problems. A lot of solutions require way more code to implement than to describe the necessary properties of. That is where you can gain some huge productivity gains by being able to focus on declaring the properties over having to define the full implementation.

But, again, it is not a panacea. No such thing exists. Every abstraction brings its own set of tradeoffs. Your job is to find the tradeoffs you can accept for your unique circumstances. What others are doing is irrelevant to your situation, but it remains that others are doing things and it can be fun to learn about it.

Re: AI demands more engineering discipline. Not less

#216

Post by ChrisLTD »

win311fwg wrote: post #213
ChrisLTD wrote: post #212

Earlier quoted context omitted.

I can be 99.99999999% certain when I write an if statement like "if (x > 1) do y" that the compiler will turn that into the equivalent machine code. So, yes, unless I hit some crazy performance bottleneck, I'm not concerned about reviewing the machine code. However, LLM outputs change with slight re-wording of prompts and with each new model release. I could hand write a test that says if x > 1 make sure y happens, b…

Just like you can be 99.99999999% certain that your properties will eventually be satisfied when the fancy compiler turns it into equivalent code. If your program is truly nothing more than that "if" statement then there is unlikely to be any productivity gain, just as there is no real productivity gain using a programming language over flipping toggle switches for something so simple. Programming languages would hav…

You're over-indexing on the (erroneous) idea that my example meant that the program would be a single if statement.

> A lot of solutions require way more code to implement than to describe the necessary properties of.

That's true to an extent, the additional code often define the emergent and undiscovered properties of a system.

Re: AI demands more engineering discipline. Not less

#217

Post by arsmoriendi »

gavinh wrote: post #101
trjordan wrote: post #27

> Those are not code problems. They are evaluation problems. > Code becomes precious when it is the only place knowledge lives. Reading AI code all day is _agonizing_. Just, a horrible way to live, and it melts people's brains at the moment you need them to be the most capable. Manual programming has this really productive and gratifying feedback loop, where you read the code, write the code, and fix it until it comp…

I agree that reading AI code all day is agonizing. We're relying on code review to develop parts of our mental model of the system that were previously developed through coding. We're having more difficulty comprehending and recall details of the system. This is probably unsurprising; people recall information better that they "generated" than information they read. I am applying some lessons from pedagogy to extend…

I also would like to know more about those pedagogy lessons you're applying.

Re: AI demands more engineering discipline. Not less

#218

Post by kazinator »

boomlinde wrote: post #186
kazinator wrote: post #169

Earlier quoted context omitted.

I understand that unskilled people are creating problems for themselves and others; but that doesn't interest me. I interpret everything from the point of view of the AI being used by skilled people. Such as the claim "AI demands more engineering discipline, not less" ... of skilled people, not irrelevant unskilled people.

Unskilled people are not irrelevant if you intend on working with other people, in which case you will need to deal with them in one way or another. I will probably be dealing with them in the future more than in the past because chatbots bypass the process by which unskilled people generally become skilled people.

Suppose I don't use AI, but have to work with people who are using it. Is AI demanding more/less engineering discipline of me?

Re: AI demands more engineering discipline. Not less

#219

Post by win311fwg »

ChrisLTD wrote: post #216
win311fwg wrote: post #213

Earlier quoted context omitted.

Just like you can be 99.99999999% certain that your properties will eventually be satisfied when the fancy compiler turns it into equivalent code. If your program is truly nothing more than that "if" statement then there is unlikely to be any productivity gain, just as there is no real productivity gain using a programming language over flipping toggle switches for something so simple. Programming languages would hav…

You're over-indexing on the (erroneous) idea that my example meant that the program would be a single if statement. > A lot of solutions require way more code to implement than to describe the necessary properties of. That's true to an extent, the additional code often define the emergent and undiscovered properties of a system.

> the additional code often define the emergent and undiscovered properties of a system.

That's the cost of abstraction. Everything has tradeoffs.

Re: AI demands more engineering discipline. Not less

#220

Post by xnx »

ryandvm wrote: post #24

It is now significantly harder to figure out who understands the systems and is using AI effectively and who doesn't know shit and is just slinging LLM copypasta around. Before 2025, the underperformers/coasters were at least relatively identifiable by the paucity of their contributions. Now all of the sudden every single engineer is filing PRs, code reviews, technical design documents, and every other artifact under…

> the fallout from this phase of the industry is going to be an exotic form of technical debt

This is a common sentiment and possibly motivated by the belief that writing "clean" code was right all along. Even odds that an AI 2 major model versions smarter will be able to eliminate that debt or rewrite from specs derived from the codebase.

Post reply on HN