The most illuminating thing about ChatGPT for me is just how terrible most programmers on HN apparently are. I thought it was just a funny meme that all we do is copy/paste from Stack Overflow but apparently that is literally what a lot of people are doing all day.
I was working on a personal project yesterday to answer some questions I had about how liquidity risk works for money market mutual funds, and to forecast/nowcast liquidity risk and NAV risk for a bunch of funds.
Mind you: I don't know the first thing about anything financial. I was just curious.
chatgpt gave me a bunch of sources of data that I wanted, translating my lay description of things I wanted to know into financial terms of art. I could then look up legal definitions and formulas for those terms to make sure they were what I thought they were. chatgpt also told me which SEC forms those things are disclosed on, what data brokers I could use for other data, etc.
between chatgpt and copilot I saved at least an hour on the job of pulling down historical data from EDGAR for a bunch of funds and getting the stats I wanted (I didn't know EDGAR existed until yesterday, and the xml/html/txt formats are kind of annoying... like, fine, but a bit of a pita so I'm glad I had help because ughhh is that kind of code boring and damn are LLMs good few-shot inductive parser generators!). Also wrote some nice chart.js code for me and helped with automatically collecting, searching, and extracting some key stats and terms from prospectuses. I didn't know about chart.js until yesterday.
All of this would've been possible without assistants, and required a lot of "executive function" on my part to bring together, but it seriously saved me at least a couple hours of implementation work and up to a day on research and learning terminology and regulatory stuff. Again, verification of those things is way easier when you know what words to look up definitions for. And chatgpt did make mistakes/hallucinate.
I don't find much use in my professional life, where the code I'm writing is apparently too domain-specific for copilot to be helpful and the mathematics is too complex for chatgpt to help with. Maybe in a few years. We'll see.
People who blindly copy from ChatGPT or Stack Overflow are most likely either very inexperienced or simply bad programmers. However, from where I see it, ChatGPT, Copilot, and any other such tools are fantastic at prompting you to think differently, getting you past writer's block, giving you ideas, or saving you some time googling for syntax yet again. You need a nudge to trigger recall of what you already know. It's a fantastic tool, just like an IDE.
It's sad how most programmers are so terrible they need IDEs to automatically create stubs for required methods or getters and setters. They should have to type all of that out every time, and from memory, too! VIM with no plugins is the only way to do it!
Or, you know, smart developers use tools appropriate to the work, and some of us have figured out how to use this new tool before you have. That's okay, you can catch up!
I don't think I'm so smart, what I am saying is the GP is being extremely dismissive of other engineers here on Hackernews. Every post about LLMs is full of engineers stating how these have been useful to them. GP simply asserts that these are all "terrible engineers." Do we think that's REALLY the case or in fact LLMs can be very useful, especially when for example learning a new language or framework they haven't used before.
Example: recently I was curious about cross-platform UI development so I used ChatGPT for research into different frameworks and made several apps in Java and .Net. Not being an engineer who's built many GUIs before, ChatGPT was amazingly useful for this and help speed my progress. Does this make me a terrible engineer?
some engineers != all engineers, that's classic unrepresentative sample. nice it helps you, hope you won't forget how to code in future, or you will be handicapped in case of openapi outage.
