211–220 of 471 posts
You got to know when to Ship it, Know when to Re-prompt, Know when to Clear the Context, And know when to RLHF. You never trust the Output, When you’re staring at the diff view, There’ll (not) be time enough for Fixing, When the Tokens are all spent.
> When you’re staring at the diff view, Bold assumption that people are looking at the diffs at all. They leave that for their coworkers agents.
Earlier quoted context omitted.
You (in theory) have more control over the quality of the team you are managing, than the quality of the models you are using. And the quality of code models puts out is, in general, well below the average output of a professional developer. It is however much faster, which makes the gambling loop feel better. Buying and holding a stock for a few months doesn't feel the same as playing a slot machine.
What theory is that? My experience is the absolute opposite. I am much more in control of quality with Ai agents. I am never letting junior to midlevels into my team again. In fact, I am not sure I will allow any form of manual programming in a year or so.
Exactly. You control the quality of the people in your team. You can train, fire, hire, etc until you get the skill level you want.
You have effectively no control over the quality of the output from an LLM. You get what the frontier labs give you and must work with that.
I think this article makes a valid point. However, if AI coding is considered gambling, then being a project manager overseeing multiple developers could also be seen as a form of gambling to a certain degree. In reality, there isn't much difference between the two. AI models are non-deterministic, and humans are also non-deterministic. You could assign the same task to two different developers and end up with entire…
I think the addiction angle seems to make AI coding more similar to gambling. Some people seem to be disturbingly addicted to agentic coding. Much more so than traditional programming. To the point of doing destructive things like waking up in the middle of the night to check agents. Or giving an agent access to their bank account.
I don’t think I’d describe my behavior as destructive though
I think this article makes a valid point. However, if AI coding is considered gambling, then being a project manager overseeing multiple developers could also be seen as a form of gambling to a certain degree. In reality, there isn't much difference between the two. AI models are non-deterministic, and humans are also non-deterministic. You could assign the same task to two different developers and end up with entire…
> AI models are non-deterministic, and humans are also non-deterministic. You could assign the same task to two different developers and end up with entirely different results. Except, one can explain themselves (humans) and their actions can be held to account in the case of any legal issue whereas an AI cannot; making such an entity completely unsuitable for high risk situations. This typical AI booster comparison…
You "own" the software it creates which means you're responsible for it. If you use AI to commit crimes you'll go to jail, not the AI.
Earlier quoted context omitted.
Only if your AI coding approach is the slot machine approach. I've ended up with a process that produces very, very high quality outputs. Often needing little to no correct from me. I think of it like an Age of Empires map. If you go into battle surrounded by undiscovered parts of the map, you're in for a rude surprise. Winning a battle means having clarity on both the battle itself and risks next to the battle.
Until it produces predictable output, it's gambling. But it can't produce predictable output because it's a non-deterministic tool. What you're describing is increasing your odds while gambling, not that it's not gambling. Card counting also increases your odds while gambling, but it doesn't make it not gambling.
Earlier quoted context omitted.
> but I finally feel like I'm _good_ at programming, which is insane Yes, it is insane. You couldn't torture this confession out of me. But that's the drug they're selling you, isn't it? You don't even write code, but you're getting a self-inflated sense of worth. It must be addicting! Of course, whether or not the programs you prompt are actually good surely has no relation to whether you feel they're good, since yo…
> Anyone can be an "ideas guy". I disagree with this. I've worked with amazing "ideas guys" who just cranked out customer insights and interesting concepts, and I've worked with lousy ones, who just kinda meandered and never had a focused vision beyond a milquetoast copy of the last thing they saw. There's a real skill to forming good concepts, and it's not a skill everyone has!
I find that the strength of people's ideas tends to be highly correlated with their overall skills. I don't know that you can develop the capability for good ideas without getting your hands dirty learning a field, experimenting, absorbing all kinds of information and understanding what really goes into the making of a good idea. In that way, the person with good ideas always ends up being more than just a ideas guy. They don't just have good ideas, they have good ideas and the skills to back them up. Whereas the "ideas guy" label is usually applied to people who have nothing to bring to the table other than their ideas, and wouldn't you know it, they aren't nearly as good as they think they are.
You got to know when to Ship it, Know when to Re-prompt, Know when to Clear the Context, And know when to RLHF. You never trust the Output, When you’re staring at the diff view, There’ll (not) be time enough for Fixing, When the Tokens are all spent.
I really hope that was your creativity and not AI
[1]: https://web.archive.org/web/20230130060050/https://www.rolli...
Earlier quoted context omitted.
One size never fits all. I am old enough to remember what a game changer Spreadsheets (VisiCalc) where. They made the personal computer into a SwissArmy knife for many people that could not justify investing large sums of money into software to solve a niche problem. Until that time PCs simply were not a big thing. I believe AI will do something similar for programming. The level of complexity in modern apps is high…
I remember when my dev team included some people using Emacs, some using Eclipse (this was pre-VS Code), and some using IntelliJ. Developers will always disagree on the best tool for X ... but we should all fear the Luddites who refuse to even try new tools, like AI. That personality type doesn't at all mesh with my idea of a "good programmer".
Earlier quoted context omitted.
Only if your AI coding approach is the slot machine approach. I've ended up with a process that produces very, very high quality outputs. Often needing little to no correct from me. I think of it like an Age of Empires map. If you go into battle surrounded by undiscovered parts of the map, you're in for a rude surprise. Winning a battle means having clarity on both the battle itself and risks next to the battle.
Until it produces predictable output, it's gambling. But it can't produce predictable output because it's a non-deterministic tool. What you're describing is increasing your odds while gambling, not that it's not gambling. Card counting also increases your odds while gambling, but it doesn't make it not gambling.
The most obvious issue is it’d class working with humans as gambling. Fine if you want to make that as your definition but it seems unhelpful to the discussion.
It is and will always be about: 1) properly defining the spec 2) ensuring the implementation satisfies said spec
This seems like 1980's corporate waterfall thinking, doesn't jibe with the messy reality I've seen with customers, unclear ideas, changing market and technical environments, the need for iteration and experimentation, mid-course correction, etc.