Live data from Hacker News

Command line functions around OpenAI

kadekillary.work

71–80 of 316 posts

Re: Command line functions around OpenAI

#71
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…

Even further, the big names in software engineering will likely figure out how to use these models far more effectively than any old engineer.

Fabrice Bellard is utilizing transformer models for lossless compression, for example: https://bellard.org/nncp/

Re: Command line functions around OpenAI

#72
post #8

What the articles like this miss is that we understand pretty well how ChatGPT writes code. But what about maintenance, and particularly debugging? How would it make these tasks. And we know that these things take much much more time than writing code.

Which is why I still believe proper developers will still have jobs :) GPT is just Spreadsheets in a way. Lot of companies had business users and SMEs create contraptions based on Excel and then when it's a hairy mess or a rube goldberg machine, there are "digital transformation" projects that are done to make them proper apps.

That might be the exact right analogy. People who see Excel for the first time seem to think they have superpowers. In actual fact the superpower is knowing how to use it effectively and knowing when you are on the edge of its abilities.

Re: Command line functions around OpenAI

#76
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…

Maybe I'm misunderstanding your last sentence - but if churning out boilerplate is something any old engineer can now have an AI do, won't it up the chances that someone previously overlooked makes a brilliant call about what problem to tackle and actually has the resources to execute on that call?

I'm doubtful. Being effective as dev usually involves a conscious effort to focus on the context of where we are, where we're going, and what's needed to get there. Also, the ability to recognize and avoid rabbit holes.

Re: Command line functions around OpenAI

#77
50c had a mil' in the bank and was still on the grind. Increased productivity doesn't mean freedom from more work.

This article is beautiful and effective. Thank you for the information!

The culture around GPT at the moment is verging on fanaticism.

When car companies close down and people lose jobs, it's sad for the die-hard followers, but most people don't care so much. This closure of trivium-level programmery is being met with a ramp up in fanaticism and repetitive doubling-down on preset views. So odd...

Re: Command line functions around OpenAI

#79
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…

> Either ChatGTP can do the whole thing itself (not yet the case) or it is a minor productivity boost because the hard part is articulating the problem.

What a ridiculous dichotomy, or course there's plenty of room in between. The hard part might be articulating the problem, but the other part takes time. Make that part not take time and you can do more of the hard part.

Re: Command line functions around OpenAI

#80
10x engineer obviously do not do 10x work, because 10x work usually means 10x bugs.

10x engineer cut the implementation complexity from 10x -> 1x, and then proceed to do that, while still satisfying business/product intention.

Generating 1000x code is more like being a -1000x engineer.

Post reply on HN