711–720 of 1001 posts
You will still need hardware to run those open models, and that avenue is far easier to contain and close than stopping code distribution. Expect the war on private/personal compute to ramp up even more significantly than ot already has.
Earlier quoted context omitted.
> non-trivial coding tasks I’ve come back to the idea LLMs are super search engines. If you ask it a narrow, specific question, with one answer, you may well get the answer. For the “non-trivial” questions, there always will be multiple answers, and you’ll get from the LLM all of these depending on the precise words you use to prompt it. You won’t get the best answer, and in a complex scenario requiring highly recurs…
>I’ve come back to the idea LLMs are super search engines. Yes! This is exactly what it is. A search engine with a lossy-compressed dataset of most public human knowledge, which can return the results in natural language. This is the realization that will pop the AI bubble if the public could ever bring themselves to ponder it en masse. Is such a thing useful? Hell yes! Is such a thing intellegent? Certainly NO!
A proofreader would have caught this humorous gaffe. In fact, one just did.
I don't understand the stance that AI currently is able to automate away non-trivial coding tasks. I've tried this consistently since GPT 3.5 came out, with every single SOTA model up to GPT 5.1 Codex Max and Opus 4.5. Every single time, I get something that works, yes, but then when I start self-reviewing the code, preparing to submit it to coworkers, I end up rewriting about 70% of the thing. So many important deta…
Code quality is an issue you need to ignore with vibe coding - if code quality is important to your project or you then it’s not an issue. But if you abandon this concept and build things small enough or modular enough then speed gains await!
IMO codebases can be architected for LLMs to work better in them, but this is harder in brownfield apps.
Seriously? If these were open source tools that anyone could run on their home PC that statement would make sense, but that's not what we are talking about here. LLMs are tools that cost massive amounts of money to operate, apparently. The tool goes away if the money goes away. Fossil fuels revolutionized the world, but only because the cost benefit made sense (at least in the relative short-term).
Efficient markets route around bottlenecks. Technological revolutions accelerate the speed at which that re-routing happens. In software, we, the developers , have increasingly been a bottleneck. The world needs WAY more software than we can economically provide, and at long last a technology has arrived that will help route around us for the benefit of humanity. Here's an excellent Casey Handmer quote from a recent…
Is that really true? I'm getting the impression that most software reinvents the wheel.
AI tools in their current form or another will definitely change software engineering, I personally think for the best However I can’t help but notice some things that look weird/amusing: - The exact time that many programmers were enlightened about the AI capabilities and the frequency of their posts. - The uniform language they use in these posts. Grandiose adjectives, standard phrases like ‘it seems to me’ - And m…
I attribute that to the holidays. Many people finally had the time to goof around with these tools. At least that's how it happened to me.
It was an incredible experience. I implemented a few features quickly and in a much better way than I could otherwise. Realized how many tiny holes my app had and a few suboptimal patterns I was using. Made me worry about my career, initially, but after using for a while, I now see it as going up the chain of abstraction. Only thing I'm not doing is writing code by hand. Im still having to do everything else like thinking about architecture and the big picture, keeping it dry and maintainable, debugging, etc - but with a lot of help from LLMs. Sometimes it's 10x and sometimes you wasted sometime, you know, just like how using packages made us go up the chain.
Earlier quoted context omitted.
This is a pretty common position: "I don't worry about getting left behind - it will only take a few weeks to catch up again". I don't think that's true. I'm really good at getting great results out of coding agents and LLMs. I've also been using LLMs for code on an almost daily basis since ChatGPT's release on November 30th 2022. That's more than three years ago now. Meanwhile I see a constant flow of complaints fro…
Show me what you've made with AI? What's the impressive thing that can convince me it's equivalent, or better than anything created before, or without it? I understand you've produced a lot of things, and that your clout (which depends on the AI ferver) is based largely because of how refined a workflow you've invented. But I want to see the product, rather than the hype. Make me say; I wish I was good enough to crea…
Stay in plan mode most of the time. It will produce a step by step set of instructions - more context - for the LLM to execute the change. It’s the best place to exert detailed control over what will happen. Claude lets you edit it in a vim window.
Think about testing strategy carefully. Connecting the feedback back into the LLM is what makes a lot of the magic happen. But it requires thought or the LLM might cheat or you get a suboptimal result.
Then with these two you spend your time thinking in terms of product correctness - good tests - and implementation plan - deciding if the LLM has a sane grasp of the problem and will create a sane result.
You’re at a higher level of abstraction, still caring about details, but rarely finicky up to your elbows in line by line code.
If you can get good at these you’re well on your way.
Earlier quoted context omitted.
On the subpar code, would the code work, albeit suboptimally? I think part of the problem a lot of senior devs are having is that they see what they do as an artisanal craft. The rest of the world just sees the code as a means to an end. I don't care how elegantly my toaster was crafted as long as it toasts the bread and doesn't break.
There is some truth to your point but you might want to consider that often seniors concerned with code quality aren't being pedantic about artisanal craft they are worried about the consequences of bad code... - it becomes brittle and rigid (can't change it, can't add to it) - it becomes buggy and impossible to fix one bug without creating another - it becomes harder to tell what it's doing - plus it can be ineffici…
* You have made a new file format. Consider that it will live forever.
* You have added exactly what the user/product team asked for it. It must be supported forever.
Part of my job is to push back on user requests. I also think a lot about ease of use.
I think even with an LLM that can one-shot a task, the engineer writing the prompt must still have "engineering judgment".
I don't understand the stance that AI currently is able to automate away non-trivial coding tasks. I've tried this consistently since GPT 3.5 came out, with every single SOTA model up to GPT 5.1 Codex Max and Opus 4.5. Every single time, I get something that works, yes, but then when I start self-reviewing the code, preparing to submit it to coworkers, I end up rewriting about 70% of the thing. So many important deta…
> I end up rewriting about 70% of the thing. Doesn't match my experience, that figure is closer to about 20-40% to me, though a lot of those changes I want are possible by just further prompting OR turning to a different model, or adding some automated checks that promptly fail and the AI can do a few more loops of fixes. > Other people are just less picky than I am, or they have a less thorough review culture that l…
…and then add that to your CLAUDE.md, and never worry about having to say it again manually.