Rendered at 12:10:13 GMT+0000 (Coordinated Universal Time) with Cloudflare Workers.
theParadox42 5 hours ago [-]
I certainly feel like certain AI coding practices I have leave me with nothing learned in a frustrating way. Sometimes the most effective way to getting what I want is just describing a goal really clearly, and then bashing the AI until it gets there. That bashing process can take a lot of energy and yet I don’t gain useful skills or a “better way” that I would if I was still coding. I generate more training data for BigAI corp, but that’s a slow return cycle.
Not sure if this is the right solution, but I would like to learn/understand/grow more as I pursue reckless efficiency.
liendolucas 4 hours ago [-]
Reason #1 to keep writing code by hand is that you don't ever lose that connection between you and the code.
There was an excellent post here on HN that unfortunately went unnoticed. I highly recommend reading it as it exactly points out this problem: understanding.
Note: The post has now been flagged. Why? I have no idea but it seems incorrect.
layer8 40 minutes ago [-]
Flying monkeys, probably. You can email hn@ycombinator.com about unwarranted flagging.
0c3ca83 2 hours ago [-]
Because it's not glazing LLMs.
0c3ca83 2 hours ago [-]
Learning is the bottleneck for shipping code. Building the model in your head, feeling out the edge cases and system interactions, and understanding the intricacies. Typing out the code is a small part of the process.
So, if you want to speed things up with LLMs, the bulk of the speed comes from reducing learning.
bootsmann 2 hours ago [-]
Imo if you’re reprompting you’re losing. If the agent oneshots the task its good but if it doesn’t manage it on the first try you will get there faster by doing it yourself.
rootnod3 4 hours ago [-]
Code by hand? The good old fashioned way? Might be "less efficient", but you will understand what you did and build experience.
freshnode 6 hours ago [-]
Have you ever heard the idiom "ambulance at the bottom of the cliff"?
pjmlp 28 minutes ago [-]
The gimmicks we come around to justify existing practices being pushed into the industry never stops surprising me.
One would expect the actual answer would be to write the code ourselves.
maximg68 2 hours ago [-]
After noticing a drop in my understanding of the code I ship, I built and started using a similar skill ("Quiz me on the PR or the current branch"). I didn't tie it to a specific commit, though - to me it makes more sense to look at the PR as a whole.
So far I was positively surprised with the questions it could come up with.
jamienk 7 hours ago [-]
I can kinda understand the motivation here: you push a bunch of slop (or is it deep, genius code?) but you have a kind of nagging suspicion that you should get to know it more familiarly, like back in the good old days a few months back when you sort of understood what was going on. I... don't... think this is going to work...? Remember Confirmation Fatigue? Soon we will have agents that interact with Sphinx for us.
chacha-bong 5 hours ago [-]
Lol I built this and immediately got frustrated with it and stopped using it. I think I don't like being questioned by someone who I expect to work like magic.
1 hours ago [-]
nkmnz 6 hours ago [-]
It’s ironic that all negative comments on here are dismissive because “slop’s baaaaad”. Yes, slop is bad. Carelessness is bad. Lack of understanding is bad. It always has been. Code has been copied from stackoverflow. Built-in methods or std libraries have been used in a wrong way. Edge cases haven’t been covered. Rate-limiting and other hardening tasks haven’t been taken care of until the system broke. The code wasn’t DRY.
Now, people are building tools that not only hide the solutions to these problems in some obscure coding tool, markdown file, or agent harness, but they build gates that help developers to elevate themselves from the darkness of not knowing better. The number of developers roughly doubles every ~3 years, so at any given time, half of the people building software have less than 3 years of experience. Most of them didn’t even study CS before starting their careers. Tools like this will be the new warning voice that told people like us 5 to 10 years ago to never copy code from stackoverflow that you didn’t understand. It’s great.
xiaoyu2006 5 hours ago [-]
The problem with AI SE is that AI writes un-architecture-ed code. The problem #2 is that a lot of people suck even more....
jdthedisciple 4 hours ago [-]
> The problem with AI SE is that AI writes un-architecture-ed code
Not necessarily. I use SOTA models with established stacks like .NET which allow my agent to easily follow decades of best practiced patterns (DI, EF Core, ASP.NET Core...) so you still and up with proper architecture.
In fact I can easily still manually inspect everything bc I already know how it's supposed to be written.
marijnz 1 hours ago [-]
From my experience, the latest models still struggle to create good architecture in the sense of good system design. You mention a couple of generally helpful patterns for the structure of the code. But they don't help much to create logical units, with intuitive boundaries, good cohesion etc. - at which even the latest models constantly make awkward decisions.
0c3ca83 2 hours ago [-]
The fact that you can say this with a straight face impresses me.
gandhiandy69 7 hours ago [-]
We went from micromanaging code to micromanaging tools and markdowns.
sshah03 6 hours ago [-]
I considered doing this with an open source app I have that let's you get a visual summary on a PR + get quizzed on it, but ultimately decided it's better to keep it an optional thing for people that want to understand AI PRs rather than force it as a gate, the volume is slowly becoming too high for even that. I agree with the sentiment behind this though as I felt this was needed as well.
alyssassan 3 hours ago [-]
Maybe the useful distinction is understanding what a PR does versus knowing how to tell if it’s wrong. A quiz can check the first, but the second is often where review matters most.
I’d be interested in a version that helps the reviewer identify one assumption the change depends on and a concrete way to test it. That could keep the exercise small while making it useful, especially if the reviewer spots something the AI-generated summary missed.
Keeping it optional also lets people reach for it when they’re uncertain, instead of turning it into another checkbox as PR volume grows.
thomask1995 5 hours ago [-]
I think this would be killer for OSS maintainers.
5 hours ago [-]
sailorganymede 5 hours ago [-]
Honestly I love to see people working on this problem ! I actually built a similar project myself (https://github.com/bhurghundii/vibecheck.dev) except it prevents you from merging on the PR level. I’d love to get in touch to talk about it sometimes.
Edit: I see there are a lot of people questioning the value of this. The thing is that there are teams where people are just pumping tons of AI code without understanding their work (literally take the ticket and dump it) and shipping it. culture change is the right answer but v hard but this at least slows things down and is quick to implement
maximg68 2 hours ago [-]
I also experienced pushback when I shared a similar skill with some of my colleagues (and it didn't mention "vibe" in the name). The main objection was "I am doing careful planning upfront, so I have the same level of understanding of the code as before".
Maybe they do. In my experience, the agent was going over the scenarios that I failed to consider during the planning stage, and it was net positive. I haven't tried it with a spec-driven approach.
sailorganymede 1 hours ago [-]
I get what they say but what I see or have seen is totally different. I come from an engineering culture where we got AI token leaderboards and mandates to use AI for every single step of the development process. We started running into bad bugs recently due to people just copy pasting the Jira into the prompt so I feel there is space to put some intentionality in the areas which matter.
zapkyeskrill 5 hours ago [-]
Does it check if the answers are AI generated?
sailorganymede 1 hours ago [-]
Not yet. But you’re like the 5th person to mention it so I think I’ll add that feature. There are some AI prevention stuff like you can’t copy paste which means you gotta use the visualisation.
jdthedisciple 4 hours ago [-]
Whenever I stop for a moment to think about it:
It is absolutely batsh*t crazy insane what SWE has come to in the last cple years.
And yet part of me is loving it.
man_luke 6 hours ago [-]
So who is the master of whom?
ericand 5 hours ago [-]
Humans are now test subjects in the ci pipeline
cjimesh 5 hours ago [-]
Its the right problem statement, but we normally let the code review process take care of this.
compounding-err 5 hours ago [-]
This is cute and novel, but practically useless. For those who use AI efficiently and effectively, this is useless. For those who don't, this becomes friction they don't want. Who really needs this?
inquirerGeneral 5 hours ago [-]
[dead]
hypfer 6 hours ago [-]
I mean I see the idea, but at least I use LLMs for code generation in two distinct ways.
The first is when I care about the code. That's when I actually read what it produces and force it to use my mental model of how things should be.
The second is when I just need a solution that sorta fits the requirements provided. Then I don't read the code because I really do not care.
So this hook would either ask me things I know, or it would ask me things I do not care about.
But I guess this is very workflow-dependent.
mikepurvis 5 hours ago [-]
I think this tool is meant to be something you impose on your colleagues who treat all code like the second bucket.
totallykvothe 7 hours ago [-]
"Outsource your thinking"
I'm sorry, what?
0c3ca83 2 hours ago [-]
Why do you think people want to build machines to think for them?
altmanaltman 7 hours ago [-]
I am trying to outsource my depression but the market is rough
freshnode 6 hours ago [-]
Yeah this is cooked
tripleee 7 hours ago [-]
we're just leaning into it now
jan_m_savage 5 hours ago [-]
a tool written by AI (mostly) to check code written by (possibly other) AI.
however, it still feels like a lot of work. what if AI hallucinates during the check?
hatch_q 4 hours ago [-]
And then you have review process done by AI as well. The trick is they all need to hallucinate the same thing and if they do, the initial requirements were wrong.
Sun Tzu said: If your people (AI agents) repeatedly fail to follow instructions, examine the clarity of your orders and the effectiveness of your leadership before blaming them.
People who care about the problem will definitely not be using this as a method to "understand" their code. People who don't care about this problem will continue to not care.
So, who is this for?
princejain756 3 hours ago [-]
The quiz axis is interesting because it tests the person, not the diff. The complementary failure mode is the stuff you can pass a quiz on while still missing: a leftover debug print, an assertion weakened to match the bug it was meant to catch, an import of a package that never existed. Those live in the diff, not in comprehension. I hit this auditing my own patches last week - I could explain every line and still shipped a stdin read that broke on non-blocking pipes. Different catches, probably need both.
Not sure if this is the right solution, but I would like to learn/understand/grow more as I pursue reckless efficiency.
There was an excellent post here on HN that unfortunately went unnoticed. I highly recommend reading it as it exactly points out this problem: understanding.
Here is the link: https://news.ycombinator.com/item?id=50128468
So, if you want to speed things up with LLMs, the bulk of the speed comes from reducing learning.
One would expect the actual answer would be to write the code ourselves.
So far I was positively surprised with the questions it could come up with.
Now, people are building tools that not only hide the solutions to these problems in some obscure coding tool, markdown file, or agent harness, but they build gates that help developers to elevate themselves from the darkness of not knowing better. The number of developers roughly doubles every ~3 years, so at any given time, half of the people building software have less than 3 years of experience. Most of them didn’t even study CS before starting their careers. Tools like this will be the new warning voice that told people like us 5 to 10 years ago to never copy code from stackoverflow that you didn’t understand. It’s great.
Not necessarily. I use SOTA models with established stacks like .NET which allow my agent to easily follow decades of best practiced patterns (DI, EF Core, ASP.NET Core...) so you still and up with proper architecture.
In fact I can easily still manually inspect everything bc I already know how it's supposed to be written.
I’d be interested in a version that helps the reviewer identify one assumption the change depends on and a concrete way to test it. That could keep the exercise small while making it useful, especially if the reviewer spots something the AI-generated summary missed.
Keeping it optional also lets people reach for it when they’re uncertain, instead of turning it into another checkbox as PR volume grows.
Edit: I see there are a lot of people questioning the value of this. The thing is that there are teams where people are just pumping tons of AI code without understanding their work (literally take the ticket and dump it) and shipping it. culture change is the right answer but v hard but this at least slows things down and is quick to implement
Maybe they do. In my experience, the agent was going over the scenarios that I failed to consider during the planning stage, and it was net positive. I haven't tried it with a spec-driven approach.
It is absolutely batsh*t crazy insane what SWE has come to in the last cple years.
And yet part of me is loving it.
The first is when I care about the code. That's when I actually read what it produces and force it to use my mental model of how things should be.
The second is when I just need a solution that sorta fits the requirements provided. Then I don't read the code because I really do not care.
So this hook would either ask me things I know, or it would ask me things I do not care about.
But I guess this is very workflow-dependent.
I'm sorry, what?
however, it still feels like a lot of work. what if AI hallucinates during the check?
Sun Tzu said: If your people (AI agents) repeatedly fail to follow instructions, examine the clarity of your orders and the effectiveness of your leadership before blaming them.
Your slop generator should check if a name exist before claiming it.
So, who is this for?