> The amount of software I can create is now mostly limited by ... hard thinking.
I'm a product manager and I was talking to my dev lead yesterday about this very thing. If PMs are like headlights on a car and devs are like the engine, then we're going from cars that max at 80mph to cars that push past 600mph, and we're headed toward much faster than that. The headlights need to extend much further into the future to be able to keep the car from repeatedly running into things.
That's the challenge we face. To paraphrase Ian Malcolm, we need to think beyond what we can build to consider more deeply what we should build.
Watching my GLM-4.7 subscription tackle problem after problem after problem & just get most of it right has really changed me a lot these past couple weeks, after being a enthusiastic but very careful pay per use coder (a lot on DeepSeek, because it's hella cheap). It is absolutely wild how much just works.
I do want better workflows where the AI thinking, where the transcript is captured. Being able to go back and understand what just happened is the major delay. And that cost increases day by day week by week, especially if the session where generation was done is lost.
This is my experience now too. The degree to which we are bottlenecks comes down to how good we are at finding the right balance between micromanaging the models (doesn't work well - massive maste of time; most of the issue you spend time correcting are things the models can correct themselves) vs. abandoning all oversight (also does not work well; will entrench major architectural problems that will take lots of effort to fix).
I spend a fairly significant amount of time revising agents, skills etc. to take myself out of the loop as much as possible by reviewing what has worked, and what doesn't, to let the model fix what it can fix before I have to review its code. My experience is that this time has a high ROI.
It doesn't matter if the steps I add waste lots of the models time cleaning up code I ultimately end up rejecting, because its time is cheap, and mine is not, and the cleanups also tend to make the time it takes to realise its done something stupid shorter.
Getting to a point where I'm comfortable "letting go" and letting the model write stupid code and letting the model fix it, before I even look at it, has been the hardest part for me of accelerating my AI use.
If I keep reading as Claude Code runs, the model often infuriates me and I end up starting to type messages to tell it fix something tremendously idiotic it has just done, only to have it realise and fix it before I get to pressing enter. There's no point doing that, so increasingly I put my sessions on other virtual desktops and try to forget about them while they're working.
It still does stupid stuff, but the proportion of stupid stuff I need to manually review and reject keeps dropping.
Looking through this guys GitHub he seems to have a lot of small “demo” apps, so I’m not surprised he gets a lot of value out of LLM tools.
Modern LLMs are amazing for writing small self contained tools/apps and adding isolated features to larger code bases, especially when the problem can be solved by composing existing open source libraries.
Where they fall flat is their lack of long term memory and inability to learn from mistakes and gain new insider knowledge/experience over time.
The other area they seem to fall flat is that they seem to rush to achieve their immediate goal and tick functional boxes without considering wider issues such as security, performance and maintainability. I suspect this is an artefact of the reinforcement learning process. It’s relatively easy to asses whether a functional outcome has been achieved, while assessing secondary outcomes (is this code secure, bug free, maintainable and performant) is much harder.
Posts like this reminds me of the classic "Most of What You Read on the Internet is Written by Insane People" [0].
The author loves vibe coding because... it lets them vibe code even more:
"One of my early intense projects was VibeTunnel. A terminal-multiplexer so you can code on-the-go. I poured pretty much all my time into this earlier this year, and after 2 months it was so good that I caught myself coding from my phone while out with friends… and decided that this is something I should stop, more for mental health than anything."
It's unclear whether the "all my time" here is "all my waking hours" or "all my time outside of my job, family duties, and other hobbies", but it's still a bit puzzling.
And so anyway, what is it that they want to code on the go so much?
"an AI assistant that has full access to everything on all my computers, messages, emails, home automation, cameras, lights, music, heck it can even control the temperature of my bed."
I guess everyone's free to get their kicks however they feel like, but - paying thousands of dollars in API fees to control your music and the temperature of your bed? Why is that so exciting?
I feel like he could make a much stronger point if there was more than a few demos and utilities for using AI tools (sell the shovels?) shared as the output.
Hmm so I can't really tell if this falls under "Ahead of the game" or "AI psychosis". The latter is usually accompanied by impact to quality of life, with hints of that where he talks about coding on his phone around friends (which thankfully he recognized is unhealthy).
Working with startups, I meet a LOT of people who obsessively cannot stop using LLMs. People who jump on MAX plans to produce as much as possible- and in the startup scene it's often for the worst ideas.
LLMs are slot machines- it's fun to pull the lever and see what you get. But the hard problem of knowing what is actually needed gets harder as we sift through ten-thousand almost-useful outputs.
12 comments
[ 6.0 ms ] story [ 28.1 ms ] threadI'm a product manager and I was talking to my dev lead yesterday about this very thing. If PMs are like headlights on a car and devs are like the engine, then we're going from cars that max at 80mph to cars that push past 600mph, and we're headed toward much faster than that. The headlights need to extend much further into the future to be able to keep the car from repeatedly running into things.
That's the challenge we face. To paraphrase Ian Malcolm, we need to think beyond what we can build to consider more deeply what we should build.
Quick math on the environmental impact of this assuming 18.35Wh/1000 tokens:
Total energy: 4.73GWh, equivalent of powering 450 average US homes annually
Carbon footprint: ~1822 metric tons of CO2, equivalent of driving 4.56 million miles in a gas powered car
Water consumption: 4.5 million litres, recommended daily water intake for 4000 people for a full year
Yet they're on twitter bragging...
https://x.com/steipete/status/2004675874499842535
I do want better workflows where the AI thinking, where the transcript is captured. Being able to go back and understand what just happened is the major delay. And that cost increases day by day week by week, especially if the session where generation was done is lost.
This is my experience now too. The degree to which we are bottlenecks comes down to how good we are at finding the right balance between micromanaging the models (doesn't work well - massive maste of time; most of the issue you spend time correcting are things the models can correct themselves) vs. abandoning all oversight (also does not work well; will entrench major architectural problems that will take lots of effort to fix).
I spend a fairly significant amount of time revising agents, skills etc. to take myself out of the loop as much as possible by reviewing what has worked, and what doesn't, to let the model fix what it can fix before I have to review its code. My experience is that this time has a high ROI.
It doesn't matter if the steps I add waste lots of the models time cleaning up code I ultimately end up rejecting, because its time is cheap, and mine is not, and the cleanups also tend to make the time it takes to realise its done something stupid shorter.
Getting to a point where I'm comfortable "letting go" and letting the model write stupid code and letting the model fix it, before I even look at it, has been the hardest part for me of accelerating my AI use.
If I keep reading as Claude Code runs, the model often infuriates me and I end up starting to type messages to tell it fix something tremendously idiotic it has just done, only to have it realise and fix it before I get to pressing enter. There's no point doing that, so increasingly I put my sessions on other virtual desktops and try to forget about them while they're working.
It still does stupid stuff, but the proportion of stupid stuff I need to manually review and reject keeps dropping.
Modern LLMs are amazing for writing small self contained tools/apps and adding isolated features to larger code bases, especially when the problem can be solved by composing existing open source libraries.
Where they fall flat is their lack of long term memory and inability to learn from mistakes and gain new insider knowledge/experience over time.
The other area they seem to fall flat is that they seem to rush to achieve their immediate goal and tick functional boxes without considering wider issues such as security, performance and maintainability. I suspect this is an artefact of the reinforcement learning process. It’s relatively easy to asses whether a functional outcome has been achieved, while assessing secondary outcomes (is this code secure, bug free, maintainable and performant) is much harder.
The author loves vibe coding because... it lets them vibe code even more:
"One of my early intense projects was VibeTunnel. A terminal-multiplexer so you can code on-the-go. I poured pretty much all my time into this earlier this year, and after 2 months it was so good that I caught myself coding from my phone while out with friends… and decided that this is something I should stop, more for mental health than anything."
It's unclear whether the "all my time" here is "all my waking hours" or "all my time outside of my job, family duties, and other hobbies", but it's still a bit puzzling.
And so anyway, what is it that they want to code on the go so much?
"an AI assistant that has full access to everything on all my computers, messages, emails, home automation, cameras, lights, music, heck it can even control the temperature of my bed."
I guess everyone's free to get their kicks however they feel like, but - paying thousands of dollars in API fees to control your music and the temperature of your bed? Why is that so exciting?
[0] https://www.reddit.com/r/slatestarcodex/comments/9rvroo/most...
Working with startups, I meet a LOT of people who obsessively cannot stop using LLMs. People who jump on MAX plans to produce as much as possible- and in the startup scene it's often for the worst ideas.
LLMs are slot machines- it's fun to pull the lever and see what you get. But the hard problem of knowing what is actually needed gets harder as we sift through ten-thousand almost-useful outputs.
[1] https://github.com/clawdbot/clawdbot/