Live data from Hacker News

Ask HN: How are you using GPT to be productive?

news.ycombinator.com

261–270 of 752 posts

Re: Ask HN: How are you using GPT to be productive?

#261
I'm using it to generate cover letters for job applications. I hate writing cover letters because it feels so insincere, so if I'm going to bullshit, may as well let an AI bullshit for me.

I don't use it for research or answering questions because it hallucinates far too much. Until these chat bots can reliably provide sources and quote those sources verbatim, it simply doesn't save me any time when I have to fact check everything it tells me. Same reason I don't trust these AIs to generate summaries, they often get little details wrong.

However, I've found it quite useful for "discovery", finding things I wasn't aware of before and may not show up using a search engine. Whether that be a library/package, a law/statute, products/brands (though monetization will inevitably ruin this), etc. I've found both Chatgpt and Bard will provide nice bulleted lists with a short description of each item, and I can do my own research from there.

Re: Ask HN: How are you using GPT to be productive?

#262
post #225

So much: - generate AWK to transform text - modify python script to make it multithreaded (when this worked first go I was very mindblown) - explain concepts that I half understand in clear language, eg explain sharpe ratio and show me the math - “what do people usually call” type questions. Like what are common job titles for people who do X and Y? - proof read my emails and make suggestions given a stated goal - se…

> modify python script to make it multithreaded (when this worked first go I was very mindblown) Sounds intriguing, could you elaborate? Which gpt did you use? What was the input like and what did the gpt produce?

This was on the first release of ChatGPT so I guess GPT-3.5. Pretty much like WASDx describes. In my case it was even more meta because I was writing a script that was making ChatGPT API calls. It’s an I/O network call so it was fairly easy to rewrite as a multithreaded generator loop, which it got right on the first go. Nice speedup of about 10X, I imagine right about up to the API rate limit.

Re: Ask HN: How are you using GPT to be productive?

#263

So much: - generate AWK to transform text - modify python script to make it multithreaded (when this worked first go I was very mindblown) - explain concepts that I half understand in clear language, eg explain sharpe ratio and show me the math - “what do people usually call” type questions. Like what are common job titles for people who do X and Y? - proof read my emails and make suggestions given a stated goal - se…

I'm curious what you mean by "reverse pitch understanding" in your second last point, could you elaborate?

I understand it to mean “I’ll explain something to you in my words, now you describe it back to me so I can evaluate whether the right message came across”.

Helpful exercise if you’re honing a pitch, because first listen understanding is key

Re: Ask HN: How are you using GPT to be productive?

#264
I pasted some headers (some of them very long, in multiple chunks) and asked the bot to write me a summary of what the API does, and what could be used to make it better. Answer were spot on. Free documentation! If you also paste the source, it can give you a documentation for a particular function as well. Same for rather complex Makefiles.

Otherwise I do like others, I use it as a quick stack overflow for uncommon APIs (to me). Or completely random questions, knowing the answer might be dicey.

It has replaced google at about 90%, I only use google to verify, and not all the time.

Re: Ask HN: How are you using GPT to be productive?

#265
I'm currently working on a game despite not being very familiar with Unity. When I have some feature I need explained, or running into some bug in the game, I'd usually use Google and find some forum post that hopefully has a similar problem to mine and go from there. With GPT, I can just tell it the problem and get an answer, without scouring through multiple posts or having to constantly readjust my query. Yes, the information it produces might be incorrect, but I think the (time needed to verify the solution provided) is less than the (time needed to produce a search query that gets the articles you want * time needed to read each article and see if it helps me).

Here's an example regarding grounded characters sticking to walls:

prompt: Thanks! I added walls and ground to my game, as well as detection so that if the player touches the ground, they will be in a grounded state. However, when the touch the vertical face of the ground/wall, they stick to it. Can you tell me what I'm doing wrong?

response: It seems like the issue you're encountering is related to the physics and collision detection in your game. The player is likely detecting the wall as the ground, and it's causing the grounded state to be triggered when it shouldn't be. To fix this, you can modify your ground detection code to take the angle of the surface into account.

Here's a solution using the Physics.Raycast method:

Assuming you have a LayerMask called groundLayer for ground detection, make sure the ground and walls are assigned to the appropriate layers in the Unity Editor

[...further details on using angle detection on collision, as well as accurate sample code]

I can even ask it to give me the reference for the _Physics.Raycast_ method:

response: Certainly! Here's the official Unity documentation for Physics.Raycast:

[URL to official docs filtered out by HN]

Physics.Raycast - Unity Manual The Physics.Raycast method allows you to perform raycasts in your scene to detect collisions with 3D objects.

[...further information on how to use it, including an accurate code example pulled from the docs]

This just drastically reduces time-to-solution, especially when you consider you can lazily leave out information and still get a good response.

In short, I treat it as though each session is a fine-tuned search engine and get a lot of value from it, but it doesn't completely replace forums or Youtube videos, for one it can't argue with itself to provide conflicting but valid alternate solutions - I'd have to argue with it/ask it to produce those.

Re: Ask HN: How are you using GPT to be productive?

#266
post #242

I might be in the minority here, but I'm not using any AI tools so far, probably to my detriment. I don't trust it with my data, and won't rely on such tools until I can self-host them, and they can be entirely offline. There is some progress in this space, but they're not great yet, and I don't have the resources to run them. I'm hoping that the requirements will go down, or I might just host it on a cloud provider.…

Me neither, but I think before long people like us are going to be left behind. We're like people who insist on continuing to ride horses in the age of the automobile.

Re: Ask HN: How are you using GPT to be productive?

#267
post #242

I might be in the minority here, but I'm not using any AI tools so far, probably to my detriment. I don't trust it with my data, and won't rely on such tools until I can self-host them, and they can be entirely offline. There is some progress in this space, but they're not great yet, and I don't have the resources to run them. I'm hoping that the requirements will go down, or I might just host it on a cloud provider.…

I don't use them either.

I've played around with ChatGPT and Copilot a little, and found that they often are subtly, but very confidently wrong in their output when asked to perform a programming task.

Sure you could spend ages refining the prompt etc, but its going to be faster to just write the fucking code yourself in the first place most of the time.

Then there's the privacy/security concerns...

Re: Ask HN: How are you using GPT to be productive?

#268
post #239
post #149

Earlier quoted context omitted.

I'm pretty much the same. I don't find significant productivity gains from using it - maybe because I have a specific way of doing things already. For instance, I know it's better for me to understand React/whatever framework rather than letting chatgpt write the react state/reducers/etc and all that stuff. I can definitely use it for emails and have used to simplify exec emails in my company but that's just it.

Understanding a topic and letting chatgpt do repetitive and simple boilerplate stuff are not exclusive.

I'm fascinated by people saying they use GPT for boilerplate. Whenever I find myself doing simple/repetitive stuff, I tend to stop in my tracks and make that go away. Usually following the rule of three: If I do something for a second time, I don't generalise/generate just yet. If I need it a third time, I sit down and do it. Is that unusual? That said, I am mostly working with high level languages that make this generalisation relatively easy.

Re: Ask HN: How are you using GPT to be productive?

#269
post #123

Since I use ChatGPT regularly, I decided to create my own client. I prefer to avoid third-party services that require privilege escalation like Grammarly and Copilot. I have developed distinct profiles for different tasks, each with its own system prompt and input method. After getting the hang of it, I plan to tweak the parameters as well. Here are some of the profiles: Explainer: a default, general purpose Q&A. The…

I was reluctant in using phrases like "skip prose", because I thought it would end up in a Waluigi.
Post reply on HN