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

Agentic Coding Is a Trap

larsfaye.com

Post reply on HN
Print view View original

191–200 of 403 posts

Re: Agentic Coding Is a Trap

#191

Post by throwaway041207 »

wiieee wrote: post #163
zbentley wrote: post #143

Earlier quoted context omitted.

I don’t disagree with any of that, but I think the brutal truth is that the priority of most businesses was always that approximate, slipshod, business-driven development. The human engineering process was only coincidentally a check back against the worst outcomes of that philosophy, not intentionally one.

Yeah but all these firms are going to get destroyed by firms led by people who are more disciplined and enforce rigour in thought etc that’ll be pushed through discouraging over-use of llm’s. Apple didn’t go from near bankrupt to where it is today without that discipline.

> more disciplined and enforce rigour

Eventually this will be automated as well. Discipline, rigor and correctness are not strictly human tasks.

Re: Agentic Coding Is a Trap

#192

Post by slopinthebag »

throwaway894345 wrote: post #119
slopinthebag wrote: post #82

Earlier quoted context omitted.

Yeah this is the same conclusion I have. I primarily use AI for UI code, and guess what, it's all basically mechanical drudgery anyways. Put a div here, or put a Box here, apply some style rules, etc. This shit should have been automated decades ago yet for some reason we're still writing the same stuff with a different "twist" today. Now if your career is built on writing out the same boilerplate code in its infinit…

I genuinely appreciated this comment—it made me chuckle. That said, I think there are better approaches to working with AI besides “here’s a big vague thing to work on, go write some code”. I think you have to iterate somewhat closely with the AI to write a doc describing exactly what you want the system to do and then scope out very narrow tickets and then have a separate agent do the TDD to actually produce the thi…

> don’t let a code writing agent have too much scope—just a narrowly scoped ticket

it's interesting cuz my intuition is to give the language model writing the files as much context as possible, which means all of the previous planning thread. but I also thought you should plan with a small model and implement with a large one, and the meta seems to be plan with an expensive one and delegate code output to smaller ones. so what do I know.

> The agent should make very small changes at a time and then test that everything still works.

yeah I think if it's treated like a codegen machine it's basically just outputting code as if you're using a dsl, except the dsl is natural language and the output is meant to be edited, no `// this is generated code, do not edit` headers

> I think AI done properly can genuinely make some tasks better

thank god I dont need to write html by hand anymore, what a pita

Re: Agentic Coding Is a Trap

#193

Post by socketcluster »

I agree with most points in the article but I disagree with the idea of coding as planning. Coding is a really bad way to plan; it maximizes sunk cost fallacy. People tend to code the first approach which came to their heads and they are resistant to revisit an approach once it has been transformed into code. Sometimes ideas which work well in the short term don't work well in the long term.

When you start coding as a way to scope out a problem, you're biasing yourself to think of everything in terms of abstractions which you invented for problems which you don't yet fully understand. My experience is that this distorts your own thinking; you are injecting your own biases into your learning process and locking down on decisions too early due to suck-cost bias.

Having a solution all coded-up and working after a couple of days creates the illusion that you've built something solid and maintainable and that any additional functionality needs to be added on top. Before you know it, the prototype has become the foundation.

It's like if I took you to some random country and told you to build a house and you started chopping wood and putting up the walls straight away. You might immediately have noticed that it's hot so you would put lots of windows... Good... But what you don't know yet is that this country gets hit by powerful cyclones once a year on average and your wooden house won't survive the first one. You started with the wrong material. It might work really well for the first few months until a point when it won't work at all and you'll have to rebuild the entire thing from scratch.

Re: Agentic Coding Is a Trap

#194

Post by josephg »

bigthymer wrote: post #189
ryandrake wrote: post #177

Earlier quoted context omitted.

> This is a common thing doctors complain about. Patients come in, saying they just need a prescription for some drug or other. Off topic, but this must be a USA-specific problem, where prescription drugs are actually marketed at Joe consumer. I think there is maybe one other country where this insane practice is allowed. Nowhere else are patients told to “Ask your doctor about Procrapin for your irritable bowl syndr…

While prescription drug marketing to consumers is an issue in the US, I think the actual problem in this situation is people Googling their issue then coming to the doctor with the conclusion of their investigation instead of letting the doctor do the investigation themselves.

Yeah. Especially in the era of doctor ChatGPT.

Sometimes patients will also have an expired prescription from a different doctor, and they want a top-up. Good doctors check. Just because some other doctor prescribed some drug 3 months ago doesn't mean its actually the right choice, or the right choice now.

Its not just a US thing. I have a few GP friends here in Australia. They complain about it too.

Re: Agentic Coding Is a Trap

#195

Post by CGamesPlay »

hibikir wrote: post #151
CGamesPlay wrote: post #135

Earlier quoted context omitted.

Exactly this. We need to be more precise than blanket statements like "agentic coding is a trap" and start figuring out what a "tasteful" application of agentic coding looks like. ChatGPT is destroying liberal arts curriculums because students can choose to not do anything of the thinking themselves and produce mediocre work that passes the bar. I think the same problem is showing itself with agentic coding, just wit…

On liberal arts is simply a matter of what the students want to get out of the class, vs what the teacher wants the students to do: There's a huge disconnect in goals and expectations, so there's no way for the teacher to actually win. The fact that there's such disconnect should give the departments pause. This doesn't happen at all for using agentic coding: What the programmer wants and what the boss wants are pret…

Honestly I'm not really thinking about the boss-programmer relationship, but rather the programmer-agent relationship. At best, you get what fnordpiglet is talking about, where it's a symbiotic relationship. On the other side of the coin, you get a parasitic relationship like the OP is talking about: the agent delivers results, you take credit, you fail to develop (or maintain) long-term skills, you become a non-value-adding middleman, you get replaced.

Re: Agentic Coding Is a Trap

#196

Post by jFriedensreich »

An aspect i don't hear about enough is using agents to make better abstractions instead of treating the agent as the abstraction. We could work on making abstractions that allow expressing what we are building in most concise form without any boilerplate. The reviews are enjoyable, if llm providers fail we have a chance to use the great abstractions by hand. The alternative i see everywhere: We don’t care about the bad abstractions because agents write the code. We don't review because its too much and too noisy and we rather blow our brains out than manually dig through that react and spaghetti codebase and writing all the boilerplate and cross references by hand. We have to build for resilience and future us.

Re: Agentic Coding Is a Trap

#197

Post by notepad0x90 »

slashdave wrote: post #76
notepad0x90 wrote: post #50

it's a fairly new way of doing things. I predict, in the future it will be more formalized and standardized like AGILE and SCRUM and all that boring stuff. The result of that though would be establishment of development patterns that are good practices. The rule of thumb is: An agent can write it, but a human has to understand it before it gets pushed to prod. I'm still not convinced about the doom and gloom over dev…

> like AGILE and SCRUM Yeah, likely > development patterns that are good practices. Wait, now you lost me

while those things sucked, patterns imprinted on devs minds, passed on from generation of devs to the other until that's just the way things are done. CI is a good example, as are some of the documentation practices. Lots of teams that don't use agile or scrum formally, follow those concepts because that's just how they've seen others do things in the industry. I predict, LLM work will get some overdone and abstract thing like agile/scrum that lots of people hate, and few thing is useful, but then out of it the actually useful bits and pieces become self-organizing standards of sorts.

Re: Agentic Coding Is a Trap

#198

Post by bulbar »

throwaway041207 wrote: post #133
byzantinegene wrote: post #116

Earlier quoted context omitted.

you have 35 years of experience and have already built up the learning capability and general framework to acquire new knowledge. you know how to use agentic coding as a tool to supplement your work. the juniors who start today don't have that, they overrely on agentic coding and do not know what they don't know

IMO, by the time todays juniors would have 5-10 years of expected experience, the entire field will be something different altogether. Language choice distribution will collapse (if not change altogether), whole new modalities of monitoring and progressive delivery guardrails will come into play, essentially creating a 24/7 incremental rollout of pure agentic code, correctness will be determined by a mix of language…

Just as "use code for contracts" failed for crypto currencies, "use AI output as prod" will fail for AI. Both is based on "just don't make catastrophic mistakes anymore".

You also wrongly assume that requirements can always easily expressed as natural language.

Another point: Software Engineering always starts where tooling capabilities stop. You don't get a competitive advantage by building without engineers what anybody everybody else can build without engineers.

Re: Agentic Coding Is a Trap

#199

Post by notepad0x90 »

kelnos wrote: post #70
notepad0x90 wrote: post #50

it's a fairly new way of doing things. I predict, in the future it will be more formalized and standardized like AGILE and SCRUM and all that boring stuff. The result of that though would be establishment of development patterns that are good practices. The rule of thumb is: An agent can write it, but a human has to understand it before it gets pushed to prod. I'm still not convinced about the doom and gloom over dev…

> The rule of thumb is: An agent can write it, but a human has to understand it before it gets pushed to prod. The article essentially claims that no, that line of thinking is false. If the agent writes all of it (or too much of it, where "too much" is still not well defined), then your ability to understand it will atrophy with time, and you will either a) never push to prod, because you can't understand it well eno…

> I think the article is correct.

What I said doesn't contradict the article. if what you said is true, then since a human can't understand it well enough, that approach is not good according to my rule of thumb, thus agreeing with article. I only established the litmus test.

agreed on other points you made.

Re: Agentic Coding Is a Trap

#200

Post by DauntingPear7 »

bigthymer wrote: post #189
ryandrake wrote: post #177

Earlier quoted context omitted.

> This is a common thing doctors complain about. Patients come in, saying they just need a prescription for some drug or other. Off topic, but this must be a USA-specific problem, where prescription drugs are actually marketed at Joe consumer. I think there is maybe one other country where this insane practice is allowed. Nowhere else are patients told to “Ask your doctor about Procrapin for your irritable bowl syndr…

While prescription drug marketing to consumers is an issue in the US, I think the actual problem in this situation is people Googling their issue then coming to the doctor with the conclusion of their investigation instead of letting the doctor do the investigation themselves.

i think this is large part due to the horrific healthcare system we have here in the US. I have personally witnessed family struggle to get help from doctors because they don't interact with a patient's health outside of immediate action items like filling a prescription, writing a summary, charting, writing a referral, etc. They spend 0 time looking into what could actually be wrong with them outside of their immediate knowledge/guess. I would like to think this is due to the shortage of doctors we have here (self inflicted) and the high demands we place on them. It really just sucks to be told to take some med, wait 3 months, and then have to communicate for them between doctors while filling out the same form 3 times.
Post reply on HN