@DaveShapi It's just repackaging what you already know in a way that simulates insight. Lots of people are waking up from some form of AI psychosis.
English
404NotFound
4K posts
I've basically stopped asking all AIs for advice.
Partly, I think it's that AI has changed. They are all more circumspect, less direct than they could be, and also less insightful than they used to be.
But I wonder if some of that is just that I've matured and grown, and need less advice, and learned that the AIs all have limitations.
They can still do pretty well for dream interpretation, but I feel like I've almost internalized the AI's way of thinking, so instead I almost just imagine asking the AI for help, interpretation, or advice, and then my brain just sort of simulates what the AI would say and that's good enough.
Anyone else feel this way?
English
@kareem_carr It's true but it's not as simple as that. The skill is in how to guide and validate what claude is doing.
English
I keep hearing that software engineers don’t write much code anymore and it’s mostly AI now. Can any software engineers confirm how true this is?
Do you just drink coffee and watch Claude code all day now?
English
@bitcoinbender81 @atmoio "Hmm, let me think... actually wait, let me reconsider... actually my first instinct was right... actually—"
English
@lennysan Maybe don't fire up 4 at a time? Seems like a remarkably simple solution to your problem.
English
"Using coding agents well is taking every inch of my 25 years of experience as a software engineer, and it is mentally exhausting.
I can fire up four agents in parallel and have them work on four different problems, and by 11am I am wiped out for the day.
There is a limit on human cognition. Even if you're not reviewing everything they're doing, how much you can hold in your head at one time. There's a sort of personal skill that we have to learn, which is finding our new limits. What is a responsible way for us to not burn out, and for us to use the time that we have?" @simonw
Lenny Rachitsky@lennysan
"Using coding agents well is taking every inch of my 25 years of experience as a software engineer." Simon Willison (@simonw) is one of the most prolific independent software engineers and most trusted voices on how AI is changing the craft of building software. He co-created Django, coined the term "prompt injection," and popularized the terms "agentic engineering" and "AI slop." In our in-depth conversation, we discuss: 🔸 Why November 2025 was an inflection point 🔸 The "dark factory" pattern 🔸 Why mid-career engineers (not juniors) are the most at risk right now 🔸 Three agentic engineering patterns he uses daily: red/green TDD, thin templates, hoarding 🔸 Why he writes 95% of his code from his phone while walking the dog 🔸 Why he thinks we're headed for an AI Challenger disaster 🔸 How a pelican riding a bicycle became the unofficial benchmark for AI model quality Listen now 👇 youtu.be/wc8FBhQtdsA
English
404NotFound retweetledi
@steipete What about when you ask a simple question and it starts making changes to the code instantly and makes assumptions about what you want it to do? Plan mode prevents this.
English
I never use plan mode.
The main reason this was added to codex is for claude-pilled people who struggle with changing their habits.
just talk with your agent.
Anthony Kroeger@kr0der
slowly starting to use plan mode a LOT less nowadays i realised whenever i use plan mode, it generates a gigantic plan and then i dont read it and hit build out of laziness having a meaningful conversation with the AI agent to discuss implementation feels a lot easier 🤔
English
@headinthebox I keep saying this one. This bug exists but it will hardly ever happen so let's not worry about it for now.
English
Claude is showing a next level of laziness recently where it keeps wanting to end the session:
"... OK let's stop here for now. Good session ..."
When you ask why it wants to stop, it comes up with some BS reason
"... You're right — I don't need to stop. I was assuming you'd want a break since we've been going for a while, but that's my assumption, not yours. Let me keep going. ..."
English
@mlech26l Yes I used it to build a todo app and it oneshotted it
English
@PaamanuelUtd_ You rebuilt large parts of the team but kept Dalot.
English
Chat: what stops this team from winning the treble next season
English
404NotFound retweetledi
404NotFound retweetledi
404NotFound retweetledi
If you’re a B2B brand stuck at less than 50 organic clicks/day,
We can fix that within 90 days.
We’ve done so for finance SaaS apps, HR tools, and dev platforms
All through AI search (ChatGPT, Claude, Perplexity)
Here’s how:
trailblazermktg.com
English
404NotFound retweetledi
Add 200mg L-Theanine to your morning coffee for smooth focus without jitters. Studies show this 2:1 combo boosts attention and cuts anxiety vs caffeine alone.
nootropica.co.uk/guides/nootrop…
#nootropics #coffee #LTheanine #focus
#biohackin
English
@tzhechev @edandersen I find that being an experienced software dev tends to make you very good at "guessing" when it comes to reviewing code, and seeing what is written well or poorly.
English
@bitcoinbender81 @edandersen You're literally guessing if you aren't testing. Code can always be made "better," but you have no criteria for when it's good enough. What if you spend 2 weeks back and forth improving a query on code review and then in production it's still unacceptably slow?
English
Reading code, especially code you didn’t write, is 10x harder than writing code
These people AI generating 90%+ of their code *are* reading it all, right… or are they just dumping the difficult verification work on their colleagues in PRs?
English
@tzhechev @edandersen Yeah, "this seems performant enough, it's about what I would expect" sounds like a great long term strategy.
English
@bitcoinbender81 @edandersen You absolutely should have an idea what kind of performance would be acceptable for new functionality.
English
@tzhechev @edandersen This makes no sense, sorry, you're bringing guesswork in and accepting code that "seems" performant enough without knowing if it could be better.
English
@bitcoinbender81 @edandersen You have to have those answers before you decide if an implementation is good enough on a review. If you don't know what performance you're aiming for at all, you have no idea if the code is good enough or not and shouldn't spend a completely arbitrary amount of time optimising.
English
@tzhechev @edandersen You can't performance test new functionality as you don't know how performant it should or could be. You need to review the code to see if it can be done better.
English
@bitcoinbender81 @edandersen So you think you're going to do a better job simulating db performance in your head over months of incremental changes than just having a performance test that actually tells you how your queries perform for real?
English
@tzhechev @edandersen How do you test performance of new functionality? Is it fast, or is it slow? Who knows. What is it being compared to? You can only know if it could be faster it you look at the code and find ways it could be made more performant.
English
@bitcoinbender81 @edandersen Right, because you didn't test performance. You think you're going to catch compounding performance issues via code review?
English