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

Build full “product skills” and you'll probably be fine

twitter.com

Post reply on HN
Print view View original

801–810 of 829 posts

Re: Build full “product skills” and you'll probably be fine

#801

Post by 6510 »

ghaff wrote: post #616
6510 wrote: post #576

Earlier quoted context omitted.

I learn in a different topic on HN that with a sophisticated chat bot lurkers dare ask support questions. As many developers I've often asked people how to do things. Before asking I try to find answers in google (as asking questions easily googled is lame) but if the chat channel is busy and I've already asked for help in the last 30 minutes I do more google searches. Each next question feels more expensive. Technic…

I don't think in 40-ish years I have ever asked a manager what's a good day or week for me to take off. I probably have certain constraints and I can look at my calendar. And have a general sense for good and less-good periods.

Purely out of curiosity I one time ask for a risk analysis report my employer is legally obligated to maintain and share with me (but didn't have). Safe to say the response involved a lot of emergent behavior.

With people asking a question is never just asking a question.

Like with any job there is a sub set of tasks better done by a machine.

Re: Build full “product skills” and you'll probably be fine

#802

Post by surgical_fire »

hnfong wrote: post #791
surgical_fire wrote: post #273

Earlier quoted context omitted.

The dude you replied to you gave out the sort of bad take I come here for. If anyone thinks genuinely that ChatGPT can meaningfully replace a software developer should be starting a tech business right now. You can replace basically your largest cost (staff wages, especially for expensive pesky software developers) with a $20 account that will work overtime, will never get tired, and whose performance will never fluc…

There's a difference between being able to do logical reasoning, and being able to do everything a competent white collar worker can do. For one, there's a token limit and memory limit, which limits the scale of the problems current iterations of GPT can do (and this limitation is not a limitation of logical reasoning ability). There's also (for GPT) no way for one to fine tune or train the model to work better in a…

> I'm just saying I think GPT does exhibit some logical reasoning abilities and not merely remembering statistical patterns.

I agree with most of your reply, except this bit.

I mean, generating a response through statistical language patterns is a sort of reasoning, and ChatGPT has been accurate enough to replace internet search for me quite often. But it also generates bullshit that an untrained eye would miss (because the bullshit it generated was statistically plausible).

When it get things wrong it generates some comically wrong behavior. I had one case where it looped through variations of the same wrong response - precisely because it is unable to do any kind of logical reasoning upon faulty or inaccurate data.

Re: Build full “product skills” and you'll probably be fine

#803

Post by surgical_fire »

hnfong wrote: post #790
surgical_fire wrote: post #214

Earlier quoted context omitted.

I use it on a daily basis.

I notice you said you used it, but did you actually intentionally test its reasoning abilities by giving it "known problems" and equivalently difficult (whatever that means) "novel" problems, and observe the difference in the quality of results it gave? (btw, apparently GPT-4 does much better on mathematical reasoning, I've yet to try myself though)

Yes.

It fails sometimes on known problems, not even particularly hard ones - i.e.: simple questions about the capabilities of programming libraries for which there is ample documentation available.

The thing is, by talking about its shortcomings I may be sounding dismissive. I'm not. I think ChatGPT is an amazing tool. I just recognize it's shortcomings to use it to its best capacity.

For things that don't have a known answer (i.e.: I couldn't easily find an answer online) it does generate plausible bullshit. On those cases it shines for things that don't really have any strictness requirements (for example, ask it to generate some fiction, ask it to generate a polite response to an email, etc). I used to suggest character names for a game I was going to play based on some parameters, and I loved the responses it gave me.

Re: Build full “product skills” and you'll probably be fine

#804

Post by scotty79 »

carlmr wrote: post #29

Looking back we had one CS professor who in 2007 predicted we'd all be jobless in ten years, i.e. 2017. His prediction was based on the trends he was seeing at the time. But it wasn't even AI. Instead he made this prediction because he saw the rise of no-code tools replacing software developers because managers could finally cut out the pesky "translators", i.e. software developers. I said it then and I will say it n…

If you have induction stove try to verbally describe what it does exactly. Half-way you'll get frustrated and switch to some computer language.

Re: Build full “product skills” and you'll probably be fine

#805

Post by chiefalchemist »

scaramanga wrote: post #784
chiefalchemist wrote: post #303

The super power I value - and rarely see in my peers - is the ability to hear wants and discuss them to define needs. The initial stated wants are rarely the actual needs. "But they said _____." Yes, they did. That doesn't mean they got it right. People say a lot of ambiguous things. A client with a product or feature in mind is no different.

Right. But you're missing the whole point. The allure of the AI hype-train is that you can turn around to your most talented and insightful developers and say "stop undermining me" and then get the machine to produce the nonsensical garbage that you would have had to spent days, weeks, months getting your crew to produce, all the while enduring their silent looks of judgement :)

Nah. You're missing the point. Get the prompt wrong and it doesn't natter who/what is doing the output.

These models are trained trained on the public internet. As your problems get more esoteric, AI's ability to recall answers becomes less accurate.

Maybe Prompt Master, and then QA is the future of tech?

Re: Build full “product skills” and you'll probably be fine

#806

Post by iamacyborg »

dragonelite wrote: post #795
iamacyborg wrote: post #114

Earlier quoted context omitted.

> the people on the floor know exactly what is killing their productivity The reality is that they frequently don’t.

People on the floor know exactly what is killing their productivity given they use the product for hours each day to do their data entry tasks.

Data entry? You must be thinking of a very specific scenario and worker.

Re: Build full “product skills” and you'll probably be fine

#807

Post by execveat »

ilaksh wrote: post #574
execveat wrote: post #541

Earlier quoted context omitted.

Yes, I'm subscribed to poe.com and am playing with all public models. They all suck at debugging issues with no known answers (I'm talking about typical problems every software developer, DevOps or infosec person solves every day). You need a real ability to reason and preserve context beyond inherent context window somehow (we humans do it by keeping notes, writing emails, and filing JIRA tickets). So while this doe…

If you think that the average public LLM is equivalent to ChatGPT or GPT-4 then you are completely mistaken. By a factor of say 500-10000%.

poe.com is a web interface (by Quora) to multiple LLMs. Right now it's ChatGPT, GPT-4, Claude, Claude+ as well as Sage and Dragonfly.

Re: Build full “product skills” and you'll probably be fine

#808

Post by latency-guy2 »

weatherlite wrote: post #694
latency-guy2 wrote: post #682

Earlier quoted context omitted.

There's nothing wrong with being confident with your skills. Humbleness is only seen as a virtue because people have an ego and don't like to be reminded that they are unskilled.

Let me guess, not doing great socially?

By all means, no one stops you from making guesses.

Re: Build full “product skills” and you'll probably be fine

#809

Post by sarchertech »

eldritch_4ier wrote: post #723
sarchertech wrote: post #424

Earlier quoted context omitted.

> But at that point every other human job is automated too. I’ve heard that exact same argument for software engineers. As an outside observer it sounds like you’ve switched from SE to founder, so you’re failing to apply the same standards because you (consciously or not) want to defend your current position.

How do they apply to software engineers exactly? The founder “job” is literally “do anything that needs doing in the business”. Now the job of a CEO or CTO is much more narrowly defined and I can see that seeing automation sooner, but the founder role is kinda unique in that it’s just “do everything and if you’re successful you should be constantly replacing your job with an employee”. Founders are the jobs trying th…

>That’s very unique

That's not even close to unique. Software engineers have been doing that since they existed.

>but the founder role is kinda unique in that it’s just “do everything and if you’re successful you should be constantly replacing your job with an employee”.

That's maybe the ideal of a founder, it's definitely part of the mythology, but it wasn't true for the majority of the many founders I worked with back when I was consulting for early stage startups. Based on my experience with a large sample size, the job of a founder is to convince VCs to give them money.

>How do they apply to software engineers exactly?

It's one of the definitions of the singularity. When AI can write itself, it will improve so rapidly that humans won't be able to keep up.

Re: Build full “product skills” and you'll probably be fine

#810

Post by sarchertech »

chiefalchemist wrote: post #757
weatherlite wrote: post #633

Earlier quoted context omitted.

I don't disagree we just don't know yet. I do think these 2 guys will be overworked as f** and constantly context switching to the point they barely know what they're doing anymore...hope we don't get there.

Funny enough, last week I was talking about Copilot with a colleague. Out loud, I wondered if there was a diminishing return from Copilot. That is, Copilot is great at filling in boilerplate / the obvious, and I get to focus on the more intricate parts of the solution. But can my brain operate at the higher level day after day? Week after week? TBH I have my doubts. Maybe it's going to be AI enables quality more than…

Same thing is happening with doctors. I read a post from a GP a while back talking about how it used to be that he got a wide mix of acuity/difficulty all day long. Now all the "easy" patients go to the Nurse Practitioner, and all he gets is the difficult ones.

He's still expected to put in the same amount of time, but now he has to expend brain power the entire time.

Post reply on HN