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

Ask HN: How can I get better at using AI for programming?

news.ycombinator.com

Post reply on HN
Print view View original

411–420 of 491 posts

Re: Ask HN: How can I get better at using AI for programming?

#411

Post by primaprashant »

johnfn wrote: post #15
bogtog wrote: post #10

Using voice transcription is nice for fully expressing what you want, so the model doesn't need to make guesses. I'm often voicing 500-word prompts. If you talk in a winding way that looks awkward when in text, that's fine. The model will almost certainly be able to tell what you mean. Using voice-to-text is my biggest suggestion for people who want to use AI for programming (I'm not a particularly slow typer. I can…

That's a fun idea. How do you get the transcript into Claude Code (or whatever you use)? What transcription service do you use?

I built my own open-source tool to do exactly this so that I can run something like `claude $(hns)` in my terminal and then I can start speaking, and after I'm done, claude receives the transcript and start working. See this workflow here: https://hns-cli.dev/docs/drive-coding-agents/

Re: Ask HN: How can I get better at using AI for programming?

#412

Post by ImPostingOnHN »

simonw wrote: post #360
hansmayer wrote: post #351

Earlier quoted context omitted.

> One of us is misleading people here, and I don't think it's me. Firstly, I am not the one with an LLM-influencer side-gig. Secondly - No sorry, please don't move the goalposts. You did not answer my main argument - which is - how does a "tool" which constantly change its behaviour deserve being called a tool at all? If a tailor had scissors which cut the fabric sometimes just a bit, and sometimes completely differe…

> how does a "tool" which constantly change its behaviour deserve being called a tool at all? Let's go with blast furnaces. They're definitely tools. They change over time - a team might constantly run one for twenty years but still need to monitor and adjust how they use it as the furnace itself changes behavior due to wear and tear (I think they call this "drift".) The same is true of plenty of other tools - potter…

> Let's go with blast furnaces. They're definitely tools. They change over time - a team might constantly run one for twenty years but still need to monitor and adjust how they use it as the furnace itself changes behavior due to wear and tear (I think they call this "drift".)

Now let's make the analogy more accurate: let's imagine the blast furnace often ignores the operator controls, and just did what it "wanted" instead. Additionally, there are no gauges and there is no telemetry you can trust (it might have some that can the furnace will occasionally falsify, but you won't know when it's doing that).

Let's also imagine that the blast furnace changes behavior minute-to-minute (usually in the middle of the process) between useful output, useless output (requires scrapping), and counterproductive output (requires rework which exceeds the productivity gains of using the blast furnace to begin with).

Furthermore, the only way to tell which one of those 3 options you got, is to manually inspect every detail of every piece of every output. If you don't do this, the output might leak secrets (or worse) and bankrupt your company.

Finally, the operator would be charged for usage regardless of how often the furnace actually worked. At least this part of the analogy already fits.

What a weird blast furnace! Would anyone try to use this tool in such a scenario? Not most experienced metalworkers. Maybe a few people with money to burn. In particular, those who sing the highest praises of such a tool would likely be ignorant of all these pitfalls, or have a vested interest in the tool selling.

Re: Ask HN: How can I get better at using AI for programming?

#413

Post by radio879 »

Lately, I’ve been using Claude Code skills/plugins.

One is called Superpowers (you should be able to find it by typing “claude code superpowers github”), and there are many other skills installed for all sorts of things. Gotta give props to whoever came up with the skills idea at Anthropic. It’s a smart way of handling context — instead of dumping pages of helper text into the prompt (for example, how to properly build a frontend website), it just provides some metadata and lets the agent decide whether it needs to load more.

I highly recommend spending time finding and installing lots of these skill plugins. They can make even the cheapest small AI models work way better.

I’ve been using the Superpowers plan mode: it creates a detailed plan first, then there’s an execute mode (and a few others I’m forgetting). I can give it a long, hard task — like converting a large app to another language - and if I run plan first and then execute, I can just let it run for hours and come back to something that’s basically done: fully tested, following best practices, with most bugs fixed.

There are some skills for frontend development, and now anytime I make a website, its crazy amazing looking and has all the stuff I usually forget.

Re: Ask HN: How can I get better at using AI for programming?

#414

Post by radio879 »

PufPufPuf wrote: post #220
frankdenbow wrote: post #204

Earlier quoted context omitted.

Have been using Opus 4.5 and can confirm this is how it feels, it just works.

It also works your wallet

Right now Google Antigravity has free Claude Opus 4.5, with pretty decent allowances.

I also use Github Copilot which is just $10/mo. I have to use the official copilot though, if I try to 'hack it' to work in Claude Code it burns thru all the credits too fast.

I am having a LOT of great luck using Minimax M2 in Claude Code, its very cheap, and it works so good.. its close to Sonnet in Claude Code. I use this tool called cc-switch to swap out different models for Claude Code.

Re: Ask HN: How can I get better at using AI for programming?

#415

Post by simonw »

ImPostingOnHN wrote: post #412
simonw wrote: post #360

Earlier quoted context omitted.

> how does a "tool" which constantly change its behaviour deserve being called a tool at all? Let's go with blast furnaces. They're definitely tools. They change over time - a team might constantly run one for twenty years but still need to monitor and adjust how they use it as the furnace itself changes behavior due to wear and tear (I think they call this "drift".) The same is true of plenty of other tools - potter…

> Let's go with blast furnaces. They're definitely tools. They change over time - a team might constantly run one for twenty years but still need to monitor and adjust how they use it as the furnace itself changes behavior due to wear and tear (I think they call this "drift".) Now let's make the analogy more accurate: let's imagine the blast furnace often ignores the operator controls, and just did what it "wanted" i…

You appear to be arguing that powerful, unpredictable tools like LLMs need to be run carefully with plenty of attention paid to catching their mistakes and designing systems around them (like sandboxed coding agent harnesses) that allow them to be operated productively and safely.

I couldn't agree more.

Re: Ask HN: How can I get better at using AI for programming?

#416

Post by tacitusarc »

Fargren wrote: post #132
dayjah wrote: post #121

Earlier quoted context omitted.

The Attention algo does that, it has a recency bias. Your observation is not necessarily indicative of Claude not loading CLAUDE.md. I think you may be observing context rot? How many back and forths are you into when you notice this?

That explains why it happens, but doesn't really help with the problem. The expectation I have as a pretty naive user, is that what is in the .md file should be permanently in the context. It's good to understand why this is not the case, but it's unintuitive and can lead to frustration. It's bad UX, if you ask me. I'm sure there are workarounds such as resetting the context, but the point is that god UX would mean s…

It’s not that it’s not in the context, it’s that it was injected so far back that it is deemed not so important when determining the next token.

Re: Ask HN: How can I get better at using AI for programming?

#417

Post by ImPostingOnHN »

simonw wrote: post #415
ImPostingOnHN wrote: post #412

Earlier quoted context omitted.

> Let's go with blast furnaces. They're definitely tools. They change over time - a team might constantly run one for twenty years but still need to monitor and adjust how they use it as the furnace itself changes behavior due to wear and tear (I think they call this "drift".) Now let's make the analogy more accurate: let's imagine the blast furnace often ignores the operator controls, and just did what it "wanted" i…

You appear to be arguing that powerful, unpredictable tools like LLMs need to be run carefully with plenty of attention paid to catching their mistakes and designing systems around them (like sandboxed coding agent harnesses) that allow them to be operated productively and safely. I couldn't agree more.

> You appear to be arguing that powerful, unpredictable tools like LLMs need to be run carefully with plenty of attention

I did not say that. I said that most metalworkers familiar with all the downsides (only 1 of which you are referring to here) would avoid using such an unpredictable, uncontrollable, uneconomical blast furnace entirely.

A regular blast furnace requires the user to be careful. A blast furnace which randomly does whatever it wants from minute to minute, producing bad output more often than good, including bad output that costs more to fix than the furnace cost to run, more than any cost savings, with no way to tell or meaningfully control it, is pretty useless.

Saying "be careful" using a machine with no effective observability or predictability or controls is a silly misnomer, when no amount of care will bestow the machine with them.

What other tools work this way, and are in widespread use? You mentioned horses, for example: What do you think usually happens to a deranged, rabid, syphilitic working horse which cannot effectively perform any job with any degree of reliability, and which often unpredictably acts out in dangerous and damaging ways? Is it usually kept on the job and 'run carefully'? Of course not.

Re: Ask HN: How can I get better at using AI for programming?

#418

Post by gentios »

Hi there, I am co-founder of Infinitcode.ai, from what you are describing I think Infinitcode AI can help especially the code reviewer and code explorer.

Basically our code explorer lets you visualize the repo into a tree for better understanding legacy code and then you can select branches and ask questions within the interactive chat so you can either refactor or understand it more in depth

Hope this helps

Its free for smaller teams up to 3 (but only the code reviewer)

Re: Ask HN: How can I get better at using AI for programming?

#420

Post by hansmayer »

simonw wrote: post #360
hansmayer wrote: post #351

Earlier quoted context omitted.

> One of us is misleading people here, and I don't think it's me. Firstly, I am not the one with an LLM-influencer side-gig. Secondly - No sorry, please don't move the goalposts. You did not answer my main argument - which is - how does a "tool" which constantly change its behaviour deserve being called a tool at all? If a tailor had scissors which cut the fabric sometimes just a bit, and sometimes completely differe…

> how does a "tool" which constantly change its behaviour deserve being called a tool at all? Let's go with blast furnaces. They're definitely tools. They change over time - a team might constantly run one for twenty years but still need to monitor and adjust how they use it as the furnace itself changes behavior due to wear and tear (I think they call this "drift".) The same is true of plenty of other tools - potter…

Whether its blast furnaces or carbon fiber, the wear and tear (macroscopic changes) as well as material fatigue (molecular changes) is something that will be specified by the manufacturer, within some margin of error and you pretty much know what to expect - unless you are a smartass billionaire building an improvised sub off of carbon fiber whose expiry date was long due. However, the carbon fiber or your blast furnace wont break just on their own. So it's a weak analogy and a stretch at that. Now for your experiment: it has no value because a) you and me both know if you told your LLM that their output was shit, they would immediately "agree" with you and go off to produce some other crap b) For this to be a scientifically valid experiment at all, I'd expect on the order of 10.000 repetitions, each providing exactly the same output. But also on this you and me both know already the 2nd iteration will introduce some changes. So stop fighting the obvious and repeat after me: LLMs are shit for any serious work.
Post reply on HN