Any coding agent gives you a working version of a feature in minutes. Ask again and you get another one, also working, slightly different. Working stopped being the filter.
My rule is simple: I ship the version with the fewest moving parts, because I'm the one debugging it at 2am. Speed of writing means nothing against speed of fixing.
What's your filter? Curious if anyone has a better rule than simplest one wins
No judgment, genuinely curious. How many of us asked ChatGPT for app ideas at some point, then ended up building one of them, or half-building it? The line between "my idea" and "the model's idea" got blurry this year. Where did your current project actually come from?
My Windows build has been in Microsoft's certification queue for three weeks. Three new models shipped in that time, two of them can supposedly do my whole job, and not one of them can get a human at Microsoft to press approve.
What's the longest you've been stuck on something you couldn't code your way out of?
Woke up tired like I'd worked a full shift. Same bug, looping, my brain solving it ten slightly different wrong ways until morning. Git log empty. Nothing real happened.
Part of me took it as proof I care about what I'm building. The other part knows a brain that won't clock out isn't a flex, it's a leak.
Do you get this one? And if you do, what actually makes it stop?
Saw a post from an 8-year dev this week. He said any junior with an agent now matches his output, and he feels no pride left in it. It got huge, because a lot of people quietly feel the same.
I read it from the other side. I'm a designer, not an engineer. For years "can you build this" meant "find someone technical and pray they stay interested." Usually they didn't. AI took the thing that hurt that dev and handed it to me.
Same shift. Loss from one seat, freedom from another.
I'm shipping a Mac app solo right now, as someone who can't write a backend from scratch. Two years ago that was a joke.
A lot of people shipping apps in 2026 don't call themselves developers. Curious what the real split looks like here. Upvote whichever comment fits, I'll share the results in a few days. https://slimsnap.ai/blog/who-cou...
Captured a page at 3066 x 28800, pasted it into Claude, and the model received 276 x 2600. Same shape, 11x smaller, about 99% of the pixels gone. Text that was 28px tall arrives under 3px.
That's the whole reason long screenshots get useless answers. The model isn't being lazy, it's reading a gray smear.
SlimSnap ships a Claude Code skill that auto-loads your latest screenshot so the agent reads the structured JSON without you pasting anything. That's the "magic" version of the workflow.
For everything else (Cursor, Lovable, bolt.new, Replit AI, claude.ai), the screenshot spec works but you paste the JSON into chat manually. It works, just not as smooth as the Claude Code flow.
Where do you actually paste screenshots into AI tools today? Vote in the comments, ideally with one line on what your screenshot loop looks like there. That tells me which agent gets the next auto-loader.
Don't take this as a date promise. Just trying to figure out which one to start with after the current backlog.
Quick one for the builders here. During a normal workday, where do you actually capture the most, Figma, your browser, an IDE, Linear, somewhere else? I'm trying to map where the volume is, because where people take screenshots seems to matter more than where they paste them. Curious what your day looks like.
Two ways my coding agent wastes my time. One: I paste a screenshot and it edits the wrong element on a busy screen. Two: it spins up sub-agents that wander off the task. Lately the wrong-element guess is the quieter, more frequent tax for me, but a few people swear the sub-agent drift is worse. Which one eats more of your day?
