371–380 of 475 posts
Earlier quoted context omitted.
I read one characterization which is that LLMs don't give new information (except to the user learning) but they reorganize old information.
That’s only true if you tokenize words rather than characters. Character tokenization generates new content outside the training vocabulary.
Earlier quoted context omitted.
I think the analogy/ substitution falls apart in that singing is generally not very stable or lucrative (for 99.999% of singers), so it is pretty rare to find someone singing who hates it. Much less uncommon to find people working in IT who hate the specific work of their jobs. And I think we do tend to (rightfully) look down on e.g. singers who lip-sync concerts or use autotune to sing at pitches they otherwise can'…
> And I think we do tend to (rightfully) look down on e.g. singers who lip-sync concerts or use autotune to sing at pitches they otherwise can't, nevermind how we'd react if one used AI singing instead of themselves. Autotune is de rigueur for popular music. In general, I'm not sure that I agree with looking down on people.
Some people love programming, for the sake of programming itself. They love the CS theory, they love the tooling, they love most everything about it. Other people see all that as an means to an end - and find no joy from the technical aspect of creating something. They're more interested in the end result / product, rather than the process itself. I think that if you're in group A, it can be difficult to understand g…
I recognize that and I kind of agree, but I think I don't entirely. Writing the "boring" boilerplate gives me time to think about the hard stuff while still tinkering with something. I think the effect is similar to sleeping on it or taking a walk, but without interrupting the mental cruncing that's going in my brain during a good flow. I piece together something mundane that is as uninteresting as it is mandatory, but at the same time my subconscious is thinking about the real stuff. It's easier that way because the boilerplate does actually, besides being boring, still connect to the real stuff, ultimately.
So, you're kind of working on the same problem even if you're just letting your fingers keep typing something easy. That generates nice waves of intensity for my work. My experience regarding AI tends to break this sea of subconsciousness: you need to focus on getting the AI to do the right thing which, unlike typing it yourself, is ancillary to the original problem. Maybe it's just a matter of practise and at some point I can keep my mind on the domain in question eventhough I'm working an AI instead of typing boilerplate myself.
Some people love programming, for the sake of programming itself. They love the CS theory, they love the tooling, they love most everything about it. Other people see all that as an means to an end - and find no joy from the technical aspect of creating something. They're more interested in the end result / product, rather than the process itself. I think that if you're in group A, it can be difficult to understand g…
Earlier quoted context omitted.
I've had the same experience as the person to whom you're responding. After reading your post, I have to ask: if you're putting so much effort into prompting it with specific points, correcting it often, etc., why not just write the code yourself? It sounds like you're putting a good deal of effort into prompting it. Aren't you worried that overtime you'll rely on it too much and your offhand knowledge will get worse…
Not OP, it be comes natural and doesn't take a lot of time. Anyway, if you want to, LLMs can today help with a ton of programming languages and frameworks. If you use any of the top 5 languages and it still doesn't work for you, either you're doing some esoteric work or you're doing it wrong.
My only conditions:
- It must be demonstrated by adding a feature on a bigger code base (>= 20 LOC)
- The added feature cannot be a leaf feature (means it must integrate with the rest of the system at multiple points)
- The prompting has to be less effort/faster than to type the solution in the programming language
You can chose any programming language/framework that you want. I don't care if it is Java, JavaScript, Typescript, C, Python, ... hell, I am fine with any language with or w/o a framework.
Most comments here surprise me: I am using Githubs Copilot / ChatGPT 4.0 at work with a code base which is mostly implements a basic CRUD service... and outside of small/trivial example (where the generated code is mostly okay), prompting is more often than not a total waste of time. Now, I wonder if I am just totally unable to write/refine good prompts for the LLM (as it works for smaller samples, I hope I am not to…
What we really need are more studies on the productivity and skill outcomes of using AI tools. Microsoft did one, with results that were very negative towards AI tools [1]. I would like to see more (and much larger cohort) studies along this line, whether they validate Microsoft's conclusions or oppose them.
Personally I do not find AI coding tools to be useful at all - but I have not put extensive time into developing a "skillset" to use them optimally. Mainly because I believe, similar to what the study by MS found, that they are detrimental to my critical reasoning skills. If this turns out to be wrong, I would not mind evaluating changing course on that decision - but we need more data.
1. https://www.microsoft.com/en-us/research/wp-content/uploads/...
Most comments here surprise me: I am using Githubs Copilot / ChatGPT 4.0 at work with a code base which is mostly implements a basic CRUD service... and outside of small/trivial example (where the generated code is mostly okay), prompting is more often than not a total waste of time. Now, I wonder if I am just totally unable to write/refine good prompts for the LLM (as it works for smaller samples, I hope I am not to…
Play with Cursor or Claude Code a bit and then make a decision. I am not on the this is going to replace Devs boat, but this has changed the way I code and approach things.
In my search I just found trivial examples.
My critic so far:
- Examples seem always to be creating a simple application from scratch
- Examples always use super common things (like create a blog / simple website for CRUD)
What I would love to see (see elsewhere): Adding a non trivial feature to a bigger code base. Just a youtube video/demonstration. I don't care about language/framework etc. ...
Earlier quoted context omitted.
> How if it hallucinate and gives you wrong code Then the code won't compile, or more likely your editor/IDE will say that it's invalid code. If you're using something like Cursor in agent mode, if invalid code is generated then it gets detected and the LLM keeps re-running until something is valid. > It is better to read documentations and tutorials first. I "trust" LLM's more than tutorials, there's so much garbage…
Plenty of incorrect code compiles. It is a very bad sign that people are making comments like "Then the code won't compile". As for my editor saying it is invalid..? That is just as untrustworthy as an LLM. >I "trust" LLM's more than tutorials, there's so much garbage out there. Yes, rubbish generated by AI. That is the rubbish out there. The stuff written by people is largely good.
Emphatic no.
There were heaps of rubbish being generated by people for years before the advent of AI, in the name of SEO and content marketing.
I'm actually amazed at how well LLMs work given what kind of stuff they learned from.
Earlier quoted context omitted.
I'm a classic engineer, so lots of experience with systems and breaking down problems, but probably Having LLMs like 2.5 now are total game changers. I can basically flow chart a program and have Gemini manifest it. I can break up the program into modules and keep spinning up new instances when context gets too full. The program I am currently working on is up to ~5500 LOC, probably across 10ish 2.5 instances. It's b…
I've heard a _lot_ of stories like this. What I haven't heard is stories about the deployment of said applications and the ability of the human-side author to maintain the application. I guess that's because we're in early days for LLM coding, or the people who did this aren't talking (about their presumed failures... people tend to talk about successes publicly, not the failures).
One is a automatic electronics test system that runs tests and collects measurements (50k+ readings across 8-12 channels)(GPT-4, now with a GUI and faster DB thanks to 2.5). One is a QC tool to help quickly make QC reports in our companies standard form (3.7). And the last is a GUI CAD tool for rendering and quickly working through ancient manufacturing automation scripts from the 80's/90's to bring them up to compatibility with modern automation tooling (3.7).
I personally think that there is a large gap between what programs are, and how each end user ultimately uses them. The programs are made with a vast scope, but often used narrowly by individuals. The proprietary CAD program that we were going to use originally for the old files was something like $12k/yr for a license. And it is a very powerful software package. But we just needed to do one relatively simple thing. So rather than buy the entire buffet, buy the entire restaurant, Claude was able to just make simple burger.
Would I put my name on these and sell to other companies? No. Am I confident other LLM junkies could generate similar strongly positive outcomes with bespoke narrow scope programs? Absolutely.