- Having it write Ansible tasks for me to automate setting up my infra.
- Generating test data for integration tests
- Create sample data for demos and marketing material.
Ask HN: How are you using GPT to be productive?
61–70 of 752 posts
Re: Ask HN: How are you using GPT to be productive?
#62Generally rewriting emails for clarity... but I found another neat use of GPT-4. For public APIs, I ask to make sure its aware of the api. Then I ask for endpoints. I find the endpoint I want. Then I ask it to code a request to the endpoint in language X (Ruby, Python, Elixir). It then gives me a starting point to jump off from. Thirty seconds of prompt writing saves me about 20 minutes of getting setup. Yes, I have…
Can you provide an example of what prompts would you use?
Here is a good example:
1) Use Chat GPT in GPT-4 mode. I have found GPT-3 doesn't work in the same way.
2) I ask "What APIs does EasyPost have?"
It will respond with 7+ API endpoints
3) I ask "Can you write code in Ruby for the rates API?"
It responds almost perfectly with workable code from my experience in Ruby.
4) Then I ask "Can you give me that in Elixir?"
It responds with something I think is about 90% right. I am not as familiar with it but it seems close.
I am not trying to replace myself... I am just trying to make my job easier. And this seems to do it.
Re: Ask HN: How are you using GPT to be productive?
#63Generally rewriting emails for clarity... but I found another neat use of GPT-4. For public APIs, I ask to make sure its aware of the api. Then I ask for endpoints. I find the endpoint I want. Then I ask it to code a request to the endpoint in language X (Ruby, Python, Elixir). It then gives me a starting point to jump off from. Thirty seconds of prompt writing saves me about 20 minutes of getting setup. Yes, I have…
This is exactly the kind of thing I hope LLM chatbots will be genuinely useful for. Though, how often do you find it completely hallucinating endpoints / parameters etc. ?
Great question. If you ask it for an API endpoint that is described online but isn't well documented publicly, it seems to default back to what it thinks you should do. For example, in one example, it hallucinates that you need a bearer token.
I don't know whether that is because that is a common way of doing things or whether a previous prompt responded with a bearer token... But it wasn't right.
For me, it's a leaping off point that often saves time if I ask the right question. To your point, you have to be quick to know enough about the API to deduce whether you and Chat GPT are in the same universe.
Re: Ask HN: How are you using GPT to be productive?
#64Earlier quoted context omitted.
I like your prompts for your script, I may integrate it into my workflow too. I have found that when I am editing my work, it is helpful to have the original sentence and the suggested sentence one after another. While this takes away the paragraph form, it clearly helps you tease out the "improvements" to either accept or reject. I also have GPT number the sentences. And then when I am done, I say - something like -…
That is an excellent idea! Perhaps someone should consider developing a smart diff tool for this purpose. To be honest, my work style can be a bit lazy at times, and this may reflect in my approach. My tool simply copies any changes directly to the clipboard. I then replace them with the original text and make adjustments on the same page. However, for more complex workflows, your approach is awesome. It could potent…
Oh yes, GPT-4 is much more compliant when asked to revise. GPT-3 sticks to it's guns when it thinks it is correct.
Re: Ask HN: How are you using GPT to be productive?
#65I use it for coding, with mixed results, for getting me going in the right direction.
I take everything it says with a grain of salt. Through some original queries about some people, I came to realize that a GPT is a stochastic parrot optimized for plausibility. Given a Venn diagram of plausible and reality, they have a large overlap. Our ideas of plausibility are informed by common repetitive observations of reality. So GPTs almost always sound truthful, and much of what they generate overlaps with truth; sometimes it does some interesting synthesizing.
For code hints in particular, which I hit it up for 2 or 3 times a day on whim, I find that the domain matters. Asking for how to do things in Jetpack Compose, I get all kinds of weird answers. Compose is a highly volatile space, GPT will synthesize function signature that are amalgamations with 3 year old stuff and newer. It helps refine my internet searches. Thank heavens for the kotlin/compose slack channels.
When I ask GPT for mainstream Python stuff, it does pretty well. Recently asking for help with parsing ical formats, it nailed a number of questions. As I moved into handling recurring events, it started getting weirder. It wrote plausible looking code with methods that didn’t exist but were hybrids of others. It missed letting me know there was a library specifically for recurring iCal events. When it came to dealing with isoweeks and timezones, it got all the right modules and functions, but put them together in weird ways. Like a blind person assembling a puzzle.
C code it does decently well. Elixir, it struggles with. Many of its elixir answers are very dated.
By and all, I treat it as a way of generating prompts/idea for me to pursue. It’s like having an eidetic four year old at your disposal whose read all of the code ever and can mention all kinds of things from association, but doesn’t really know. Or like interviewing someone who’s crammed on a subject, but hasn’t really experienced what they’re talking about.
Re: Ask HN: How are you using GPT to be productive?
#66Generally rewriting emails for clarity... but I found another neat use of GPT-4. For public APIs, I ask to make sure its aware of the api. Then I ask for endpoints. I find the endpoint I want. Then I ask it to code a request to the endpoint in language X (Ruby, Python, Elixir). It then gives me a starting point to jump off from. Thirty seconds of prompt writing saves me about 20 minutes of getting setup. Yes, I have…
> Generally rewriting emails for clarity...
This is the sort of thing that will force a lot of legal teams to shutdown access to GPT-4 api/gui from internal networks.
Ppl never think of unintended consequences.
Ask it a prompt fine but don't provide internal information as an input.
Re: Ask HN: How are you using GPT to be productive?
#67Earlier quoted context omitted.
Can you provide an example of what prompts would you use?
Here is a good example: 1) Use Chat GPT in GPT-4 mode. I have found GPT-3 doesn't work in the same way. 2) I ask "What APIs does EasyPost have?" It will respond with 7+ API endpoints 3) I ask "Can you write code in Ruby for the rates API?" It responds almost perfectly with workable code from my experience in Ruby. 4) Then I ask "Can you give me that in Elixir?" It responds with something I think is about 90% right. I…
Note: I tried with GPT-3.5 and it doesn't respond with all the same APIs available. That said, if you want to try the above... It appears that the rates api isn't available in 3.5 but if you follow the example through.... it will still produce nearly identical code for the rates API even though it doesn't say that it is there.
Re: Ask HN: How are you using GPT to be productive?
#68Earlier quoted context omitted.
Fairly sure it's better than nothing. Until it isn't. Live and learn, as they say.
Has there ever been a case of a ToS so badly written that not having one would have been preferable? I'd be curious to hear about that story if it exists.
At a certain point in my life, I came to the conclusion that if something is important enough, it generally pays to: (1) either research the matter myself until I was satisfied that I understood the cost/risk tradeoff sufficiently; or (2) if I don't have the time or skills to do that, have the matter reviewed professionally. Both of which are alternatives to "doing nothing".
Employment or other contracts, health decisions, taxes ... that's how I roll.
That's just me, and I'm not you. It may also just be a hobby project or otherwise of negligible consequence. In which case it would seem to fall under the rubric of what generative AI is arguably suitable ("better than nothing").
Re: Ask HN: How are you using GPT to be productive?
#69Generally rewriting emails for clarity... but I found another neat use of GPT-4. For public APIs, I ask to make sure its aware of the api. Then I ask for endpoints. I find the endpoint I want. Then I ask it to code a request to the endpoint in language X (Ruby, Python, Elixir). It then gives me a starting point to jump off from. Thirty seconds of prompt writing saves me about 20 minutes of getting setup. Yes, I have…
> Generally rewriting emails for clarity... This is the sort of thing that will force a lot of legal teams to shutdown access to GPT-4 api/gui from internal networks. Ppl never think of unintended consequences. Ask it a prompt fine but don't provide internal information as an input.
Get ready for ChatGPT: Enterprise edition! Now with SOC 2 compliance!
Re: Ask HN: How are you using GPT to be productive?
#70Me: what do you think about Robert Frost
GPT: I speak the road plausibly travelled.