Do people really try to one-shot their AI tasks? I have just started using AI to code, and I found the process very similar to regular coding… you give a detailed task, then you iterate by finding specific issues and giving the AI detailed instructions on how to fix the issues.

It works great, but I can’t imagine skipping the refinement process.

▲

sdesol 6 days ago | parent | next [-]

> Do people really try to one-shot their AI tasks?

Yes. I almost always end with "Do not generate any code unless it can help in our discussions as this is the design stage" I would say, 95% of my code for https://github.com/gitsense/chat in the last 6 months were AI generated, and I would say 80% were one shots.

It is important to note that I can easily get into the 30+ messages of back and forth before any code is generated. For complex tasks, I will literally spend an hour or two (that can span days) chatting and thinking about a problem with the LLM and I do expect the LLM to one shot them.

▲

jplusequalt 6 days ago | parent [-]

Do you feel as if your ability to code is atrophying?

▲

sdesol 6 days ago | parent [-]

Not even remotely since the 5% that I need to write is usually quite complex. I do think my writing proficiency will decrease though. However my debugging and problem solving skills should increase.

Having said all of that, I do believe AI will have a very negative affect on developers where the challenge is skill and not time. AI is implementing things that I can do if given enough time. I am literraly implementing things in months that would have taken me a year or more.

My AI search is nontrivial but it only took two months to write. I should also note the 5% that I needed to implement was the difference between throw away code and a usuable search engine.

▲

jplusequalt 5 days ago | parent [-]

>Not even remotely since the 5% that I need to write is usually quite complex.

Not sure I believe this. If you suddenly automate away 95% of any task, how could it be the case you retain 100% of your prior abilities?

>However my debugging and problem solving skills should increase

By "my", I assume you mean "my LLM"?

>I do think my writing proficiency will decrease though.

This alone is cause for concern. The ability for a human being to communicate without assistance is extremely important in an age where AI is outputting a significant fraction of all new content.

	▲	sdesol 5 days ago \| parent [-]
		> Not sure I believe this. If you suddenly automate away 95% of any task, how could it be the case you retain 100% of your prior abilities? I need to review like crazy now, so it is not like I am handing off my understanding of the problem. If anything, I learn new things from time to time, as the LLM will generate code in ways that I haven't thought of before. The AI genie is out of the bottle now and I do believe in a year or two, companies are going to start asking for conversations along with the LLM generated code, which is how I guess you can determine if people are losing their skill. When my code is fully published, I will include conversations for every feature/bug fix that is introduced. > The ability for a human being to communicate without assistance is extremely important I agree with this, but once again, it isn't like I don't have to review everything. When LLMs get much better, I think my writing skills may decline, but as it currently stands, I do find myself having to revised what the LLM writes to make it sound more natural. Everything is speculation at this point, but I am sure I will lose some skills but I also think will gain new ones by being exposed to something that I haven't thought of before. I wrote my chat app because I needed a more comfortable way to read and write long messages. For the foreseeable future, I don't see my writing proficiency to decrease in any significant manner. I can see myself being slower to write in the future though, as I find myself being very comfortable speaking to the LLM in a manner that I would not to a human. LLMs are extremely good at inferring context, so I do a lot lazy typing now to speed things up, which may turn into a bad habit.

▲

ssharp 6 days ago | parent | prev [-]

Every tool I've tinkered with that hints at one-shotting (or one-shot and then refine) ends up with a messy app that might be 60-70% of what you're looking for but since the foundation is not solid, you're never going to get the extra 30-40% of your initial prompt, let the multiples of work needed to bolt of future functionality.

Compare that to the approach you're using (which is what I'm also doing), and you're able have have AI stay much closer to what you're looking for, be less prone to damaging hallucinations, and also guide it to a foundation that's stable. The downside is that it's a lot more work. You might multiply your productivity by some single digit.

To me, that 2nd approach is much more reasonable than trying to 100x your productivity but actually end up getting less done because you end up stuck in a rabbit hole you don't know you're in and you'll never refine your way out of it.

	▲	braaileb 6 days ago \| parent [-]
		I got stuck in that rabbit hole you mention. Ended up ditching AI and just picked up a no/low-code web app builder cause I don’t handle large project contexts in my own head well enough to chunk design into tasks that AI can handle. But the builder I use can separate the backend from the front end which allows for a custom front end template source code to be consumed by an ai agent if you want. I’m hoping I can manage this context better but I still have to design and deploy a module to consume user submitted photos and process with an ai model for instant quote generation