To put it mildly this @remote rebrand caught me off-guard.
All notes
Replied to
Juanjo Valiño
bro described literally the option I hate the most
I hate this with passion
Replied to
Luke Wroblewski
and you can get it now....
steer agents across devices, review their work in browsers, before & after screen shots, diagrams, code changes, etc. run terminals, scripts, custom agents across providers. all from your phone.
https://intentapp.dev
I’ve been waiting for this! Loving Intent so far and all the constant updates. Thanks to the whole team.
Replied to
Karri Saarinen
yeah and we do:
1. keep developing the feature behind internal flag, anyone trying it will keep hitting issues
2. bring external beta users
3. keep using internally still
4. do feature roast where internally the whole team tries to be nitpicky as possible and write their feedback
5. release to more users
6. release to all and keep monitoring
7. team doesn't move on for a while even if the project is kind of completed, they will clean up flags, do some other work, while monitoring the feedback coming through from users
The “team doesn’t immediately move on after project completion” part is super important.
You need time to see how it actually lands, learn from real usage, and make those important tweaks while the context is still fresh.
Replied to
Joshua Ariza
I ALWAYS tell new or young designers not to freelance.
Learn the business and process from someone first.
Work somewhere. Learn on their dime.
Well whynotboth.gif
I was always chasing freelance work on nights and weekends. That hunger helped because I never specialized early on. Any design or web project was “sure, I can do that.” I learned on the fly, built broad skills, and got great at managing clients.
https://thecursormag.substack.com/p/the-ai-to-ui-trap
low AI adoption among designers might be self-preservation, I mean something more than protecting our job. I mean preserving the idea that design is a thinking discipline, not an output discipline.
Replied to
Matt Ström-Awn
New essay! In which I try to define the toolmarks that AI leaves behind in the process of turning your prompt into a generated output.
mattstromawn.com/writing/expa...
Great post! something to keep in mind, especially with how fragmented everything feels right now.
Everyone’s constantly trying new tools, and over time people start trusting them a bit too blindly. I get it, they help move things faster, but they also help people avoid doing “the work.”
The weirdest flex right now is “I one-shotted this with AI.”
Cool. So can a stranger who types a similar sentence.
One-shotting means the output required zero judgment. Zero editing. Zero taste. You skipped the entire part that makes it yours. The work is everything that happens after the first result.
Replied to
Eugene
added this button interaction to my portfolio
(Stole or got the inspiration somewhere)
These are the kind of things I love in personal websites 👏
Looks amazing!
Generative UI is getting hyped as the future of software and it’s starting to feel like “voice will kill screens” all over again.
If every user sees a different interface:
- How does support work?
- How do you write help docs?
People actually like standards. You learn one app and carry that muscle memory to the next. That’s not a limitation, that’s how humans work.
There’s also an assumption that users know exactly what they need and can articulate it clearly. We know that’s not true.
The sweet spot is probably 80% well-designed UI for everyday tasks, 20% AI for the long tail.
But that’s not the future of software.
That’s a good feature.
Replied to
Sebastian De Deyne
Using Codex 80% of the time but for some reason Claude has *way* better taste making UI decisions
In my experience, gemini has better UI instincts. That’s when I switch to cursor to make gemini do a UI pass for quick things. Instead of keep correcting Claude for UI changes.
1 of 27
Next Page