Live data from Hacker News

The Leverage of LLMs for Individuals

mazzzystar.github.io

121–130 of 311 posts

Re: The Leverage of LLMs for Individuals

#121
I've used ChatGPT for personal project planning. I like to use Trello for personal organization. I just prompt ChatGPT with things I'd like to work on.

For example, I've given a prompt akin to: "Can you create for me a plan for organizing my email? It should be a list of tasks to be completed in 15-30 minutes. It should follow the S.M.A.R.T. format."

Then I gave it prompts to split all the tasks into single line sentences with a new line in between.

You can copy paste this into a card and Trello will ask if you'd like to split it up into multiple cards!

Re: The Leverage of LLMs for Individuals

#122
> Staying in any company right now is a negative return; you are wasting personal leverage.

I think it's a bit sad that the only benefit someone can see in their workplace is their salary.

Also, I get the feeling this person has never done anything outside of consumer app development with minimal stakeholder interaction.

Overall I think it's great we can supercharge our learning and coding capabilities, but it would be naive to think knowledge and coding can let you do anything you want.

Re: The Leverage of LLMs for Individuals

#123
post #87

I am happy that so many people get to experience the feeling me and other software engineers feel when we write code. OpenAI democratize some of this feeling and now people that never knew how to code can experience the rush of creating something new using code. In a world where a lot of AI Researchers that develop these amazing models are slowed down by their lack of "normal" engineering skills I think it's very app…

[flagged]

Re: The Leverage of LLMs for Individuals

#124
post #43

Earlier quoted context omitted.

The beauty of abstraction, though, is that it liberates a person from needing to worry about lower-level complexities. The same should be true for LLMs. It shouldn’t need to know your entire system. Start a new thread with judiciously restricted context when you move on to the next task.

Yup. We're back to the real problems that matter in sdev -- thinking about the problem, breaking it down, scoping things. Let an LLM do the grunt-work of writing the code, sure.

This is dangerous! How do you know it's correct!? What if you need different memory management!? What if it writes the wrong thing?!

Program only in Assembly, because otherwise you risk some real dangers of not knowing what's going on with the code!

Re: The Leverage of LLMs for Individuals

#125

> Staying in any company right now is a negative return; you are wasting personal leverage. I think it's a bit sad that the only benefit someone can see in their workplace is their salary. Also, I get the feeling this person has never done anything outside of consumer app development with minimal stakeholder interaction. Overall I think it's great we can supercharge our learning and coding capabilities, but it would…

> I think it's a bit sad that the only benefit someone can see in their workplace is their salary.

It's not the only benefit, but it's the only one that you can't get in some other way. I suspect that very few people -- even people who love their jobs -- would continue working in them if they weren't getting paid.

Re: The Leverage of LLMs for Individuals

#126

Earlier quoted context omitted.

Correcting the correction of a correction: The chat interface is no longer using GPT-3.5, since May 3rd. What is the default model called? I have no idea. The release notes page [1] calls it ChatGPT model, the bottom of chat interface calls it "ChatGPT May 3 version". Also people usually call out when they use GPT-4, but in the end you can only assume and be wrong 50% of time. [1] https://help.openai.com/en/articles/…

For whatever it's worth, playing with the model selector gives you a parameter in the URL. The default model is "text-davinci-002-render-sha" The legacy one is "text-davinci-002-render-paid" GPT-4 is just "GPT-4" One other useful tell if someone is showing you a screenshot of chatGPT is that GPT-4 uses a black logo avatar, where 3.5 uses a green one.

Wait, isn't the default model of ChatGPT called "gpt-3.5-turbo"? IIRC "text-davinci-002" is the old one.

Re: The Leverage of LLMs for Individuals

#127
I really liked this article overall, but was very confused by this section -

> If you are an individual developer, the leverage it provides might be 10 times greater, but when you work for a company, that number might be only 2.

Where do these numbers come from? Did ChatGPT hallucinate them? I don't understand why both can't be true. Why can't I have 10x leverage in my job and use that productive to unlock more time and then further leverage myself as a personal developer? Very strange sentiment here and also a baseless claim IMO.

Re: The Leverage of LLMs for Individuals

#128
post #87

I am happy that so many people get to experience the feeling me and other software engineers feel when we write code. OpenAI democratize some of this feeling and now people that never knew how to code can experience the rush of creating something new using code. In a world where a lot of AI Researchers that develop these amazing models are slowed down by their lack of "normal" engineering skills I think it's very app…

People who use ChatGPT to spit out code as a method of learning software engineering are going to be seriously deficient in fundamentals like performance.

I've already worked with way too many engineers who copy/paste code without understanding what it does.

Re: The Leverage of LLMs for Individuals

#129
post #71

Earlier quoted context omitted.

Thanks for checking it out! "Today, GitHub Copilot is behind an average of 46 percent of a developers' code across all programming languages -- and in Java, that number jumps to 61 percent." - https://visualstudiomagazine.com/articles/2023/02/15/copilot...

That post just cites an original [1] from someone at GitHub which says: > Case in point: When we first launched GitHub Copilot for Individuals in June 2022, more than 27% of developers’ code files on average were generated by GitHub Copilot. Today, GitHub Copilot is behind an average of 46% of a developers’ code across all programming languages—and in Java, that number jumps to 61%. I'm not sure what this means. GitH…

Even if they were somehow measuring all code created in the world, considering it's even easier than copy & pasting general code to ask AI for shitloads of code without even bothering to review it, so code created would be quite a poor metric compared to code put into production at least once.

Re: The Leverage of LLMs for Individuals

#130

> Staying in any company right now is a negative return; you are wasting personal leverage. I think it's a bit sad that the only benefit someone can see in their workplace is their salary. Also, I get the feeling this person has never done anything outside of consumer app development with minimal stakeholder interaction. Overall I think it's great we can supercharge our learning and coding capabilities, but it would…

> I think it's a bit sad that the only benefit someone can see in their workplace is their salary. It's not the only benefit, but it's the only one that you can't get in some other way. I suspect that very few people -- even people who love their jobs -- would continue working in them if they weren't getting paid.

Well yeah, but we're unlikely to ever live in a world where there aren't paid jobs. Personally, I enjoy working with people. I may not make as much money as going solo but holistically my work life is improved by having colleagues. I think the same is true for most people. No person is an island.
Post reply on HN