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

Command line functions around OpenAI

kadekillary.work

Post reply on HN
Print view View original

121–130 of 316 posts

Re: Command line functions around OpenAI

#121

Post by packetlost »

fendy3002 wrote: post #118
roenxi wrote: post #10

Being efficient at writing code can barely make someone a 5x engineer. A team of people who write code effectively is easy to do. Hyperproductivity in software is all about deciding what problems to tackle. Richard Hipp isn't worried about chipotle restaurant orders in golang, he's worried about how to store data reliably. That isn't a coding puzzle that ChatGTP is likely to help with. Either ChatGTP can do the whole…

Uh no, ChatGPT is good as being a better (contextual) search engine, a rubber duck that explain code to you (which stil bad but keep improving), and a mid-level code generator junkies that can generate crud operation, unit tests or transpile code between language. All of it will reduce mental load for senior devs and enable them to tackle more problems or more difficult problems.

> a mid-level code generator junkies that can generate crud operation, unit tests or transpile code between language.

Not really though, at least not yet. I've seen some impressive results for trivial cases: the type of thing you'd see blogged on new web devs blog 10x over, but I've yet to see anything that wasn't trivial and worked come out of GPT.

Re: Command line functions around OpenAI

#122

Post by ravenstine »

irrational wrote: post #67
esperent wrote: post #6

> First off, you’ll need an OpenAI API key. You can get this by going here and signing up. Note that you'll also need to be from a Western country, or at least have access to both a phone number from one and a VPN. And if you want a plus subscription, you'll also need a US credit card - not stated anywhere but I've been totally unable to sign up for plus with either Irish or Belgium cards, it just immediately flashes…

Why are you trying to pay for it? Let procurement at your company figure it out. You wouldn’t pay for your IDE or any other tools, would you? The business should be paying for any tools they want me to use.

Oh, my sweet, innocent child.

Have you ever tried getting a company to fund a developer tool? More often than not, it's like pulling teeth, even if the service in question is a paltry $20 a month.

I've learned to pay for things like OpenAI myself because it's not worth the hassle to get a company to pay for it (for the purpose of R&D).

Re: Command line functions around OpenAI

#123

Post by zeroonetwothree »

roenxi wrote: post #10

Being efficient at writing code can barely make someone a 5x engineer. A team of people who write code effectively is easy to do. Hyperproductivity in software is all about deciding what problems to tackle. Richard Hipp isn't worried about chipotle restaurant orders in golang, he's worried about how to store data reliably. That isn't a coding puzzle that ChatGTP is likely to help with. Either ChatGTP can do the whole…

> A team of people who write code effectively is easy to do.

Based on my experience, you’d be surprised how hard it is sometimes. I have a team of 10 engineers that failed to ship a basic CRUD UI after several months (and it wasn’t even close to done). Eventually it got taken over by one individual eng who finished it in two weeks.

Re: Command line functions around OpenAI

#124

Post by horns4lyfe »

zeroonetwothree wrote: post #123
roenxi wrote: post #10

Being efficient at writing code can barely make someone a 5x engineer. A team of people who write code effectively is easy to do. Hyperproductivity in software is all about deciding what problems to tackle. Richard Hipp isn't worried about chipotle restaurant orders in golang, he's worried about how to store data reliably. That isn't a coding puzzle that ChatGTP is likely to help with. Either ChatGTP can do the whole…

> A team of people who write code effectively is easy to do. Based on my experience, you’d be surprised how hard it is sometimes. I have a team of 10 engineers that failed to ship a basic CRUD UI after several months (and it wasn’t even close to done). Eventually it got taken over by one individual eng who finished it in two weeks.

That sounds like it was because there were 10 engineers assigned to a way too small task, not becomes teams are ineffective.

Re: Command line functions around OpenAI

#125

Post by analognoise »

lordnacho wrote: post #48
superb-owl wrote: post #46

There seems to be a lot of dissenting opinion on how these tools will affect software engineering jobs. Those of us that have deep architectural skills, are able to deal with ambiguous requirements, balance UX with technical limitations, etc, will be safe for a good while. Those of us that were trained in bootcamps to crank out MEAN/MERN/etc apps might be in a bit of trouble. I say this not to disparage the latter gr…

Question is whether you have to go through the bootcamp stage to become an architect? Like many skills, you often start with the simple stuff.

Business isn't rational. They won't think "if we don't have someone do the rote stuff in 20 years, we will have no architects."

All they think is "right now it's cheaper so we'll do that".

Re: Command line functions around OpenAI

#126

Post by bckr »

qikInNdOutReply wrote: post #107
nicpottier wrote: post #26

Earlier quoted context omitted.

I mean saying that using ChatGPT will make you one of the most elite coders in the world is a pretty high bar no? I have been skeptical about all this, but decided to play with using ChatGPT for coding a task I had. It wasn't perfect, either in workflow or in final results but what was clear was that not adopting it will very quickly make you lag behind your peers. I think the additive bonus is especially high for se…

The tooling will be pretty interesting.. basically prompts per file and a overlaying diff per prompt.. Focus mostly on the almost functional design of software, with well defined input and output as the most important criteria. Also lots of "meta" tags, that filter the trained in network layers for "fast" code, "secure" code, "minimal working smaple" code by injecting addtional prompts. The trick is to use the prompt…

I imagine that a web service could be specified in one file. At the top, a description of the service, including the language it is to be written in and how it will be built and deployed. Then an outline of the endpoints and their functionality.

A builder bot could then ask a series of questions and get permissions such as “can we use X package?”, and then update the specification before performing the build.

Re: Command line functions around OpenAI

#127

Post by Arubis »

The 10x engineers I've met don't write 10x more or better or more well-targeted code.

They enable ten other engineers to double their output.

"10x engineers" are a leadership position, not a development position.

Re: Command line functions around OpenAI

#128

Post by nolawi89 »

Lewton wrote: post #41
nolawi89 wrote: post #21

Earlier quoted context omitted.

Not yet; I spend 60% my day reviewing PRs. I don't have to if chat GPT gets integrated as a PR reviewer..

ChatGPT will turn your day into spending 90% of your time reviewing code...

what do you mean? Do you mean that there is infinite amount of code because its also writing so much code?

Re: Command line functions around OpenAI

#129

Post by tomas789 »

My experience with code assistant is strikingly similar to my current experience with Vim. Sometimes, it makes me slightly faster. Other times the experience breaks down spectacularly on previously simple things. And last but not least, it is difficult to get rid of it.

Re: Command line functions around OpenAI

#130

Post by lolinder »

uncletaco wrote: post #35

At what point did we stop making fun of that 10x engineer thread and started taking it seriously? Who forgot that 10x was a silly little thing that made zero sense in the real world?

As far as I know the 10x number was made popular by Fred Brooks, not a meme, and it has pretty real applications:

> Programming managers have long recognized wide productivity variations between good programmers and poor ones. But the actual measured magnitudes have astounded all of us. In one of their studies, Sackman, Erikson, and Grant were measuring perfor- mances of a group of experienced programmers. Within just this group the ratios between best and worst performances averaged about 10:1 on productivity measurements and an amazing 5:1 on program speed and space measurements! In short the $20,000/year programmer may well be 10 times as productive as the $10,000/year one.

He goes on to explain that because the primary cost in software development is the overhead of coordinating minds, a company is far better off paying for just a few really good programmers than they are 10 times that number of mediocre ones. Whether or not there's such a thing as a 10x programmer in isolation, this seems pretty self-evident and valuable to keep in mind.

Post reply on HN