@thsottiaux@PhilippSpiess Is it just me or does opening the sidebar to view changed files, doesn’t always let you see all changed files? Only the most recently changed in your chat.
I’ve been working toward a modernization of Slawp(https://t.co/Cboqpp0fDC), bringing the app in line with modern 2026 iOS design and Liquid Glass. I’m starting small with a redesign and rebuild of the Image Detail view.
After struggling to get there with Fable 5, I switched to GPT-5.6 Sol.
I was able to work back and forth with GPT-5.6 Sol, refining my vision and discussing how it could be implemented while staying aligned with Apple’s design principles. Combined with the Build for iOS plugin, it helped bring that vision to life.
GPT-5.6 Sol nailed it:
- Created a gradient by sampling the image’s colors
- Added paging between images
- Navigation transitions in and out of the Grid views
- Animated the image gradient
- Built Liquid Glass action buttons and metadata cards
It handled the SwiftUI patterns and layout logic far more cleanly than Fable, and the results felt production-ready much faster.
I’m really happy with how it turned out. This is just the first step in modernizing the app. @slawpai has some big features coming up (e.g. profile and avatar updates) that I plan to build with GPT-5.6 Sol.
@tibor_tee One caveat is making sure you have strict architecture documentation, and good codebase patterns, because Composer will still go off the rails if left to its own devices.
@tibor_tee And now at this point, unless I’m using plan mode for something really big/new, I rarely need a frontier model, and go straight to Composer.
@cursor_ai if $10k credit drops are real, https://t.co/1pS3AD1PJD is ready
I’ve built our iOS app entirely in Cursor for 1.5 months: https://t.co/YuTLGx7KOm
My co-founder is still on Claude Code. Historic conversion opportunity.
Live on iOS, Android, web.
"Engineering, product, and design are all merging into a 'builder' role"
Yeah... I'm not so sure. This feels like an oversimplification and podcast talking point. Reality is a lot more complex.
Even with 1000 "Member of Technical Staff" titles, someone still has to wake up and care 100x more about Product or Design than anyone else. It is their Main Thing™
That's not to say MTS titles are universally bad, but I think they're an example of this 'builder' talking point that's become bastardized.
AI and coding agents have made generating code easy and yet... you're in for a world of pain if non-engineers ship a bunch of slop and don't have great engineers to tame the complexity.
The SF hivemind has a tendency to overfit what works at startups for every company. And to be fair, sometimes this is true! Startups can be a leading indicator for how the industry is changing and often cause disruption.
However, it is going to be incredibly hard to disrupt the extremely human parts of corporate jobs. You really think there's going to be a PM who also does some engineering and design on the side at JPMorgan Chase?
This is true for the simple parts of most jobs, like people wanting to have ownership over something and do good work, move up a career ladder, support their family, get paid well, make an honest living...
And also the hard parts: internal politics, some critical business system that has a bus factor of 1 which has been running for 15 years and isn't documented anywhere because it's that guy's job security. The real world has a lot of this stuff.
It's easy to pontificate about all roles collapsing but it's actually really nice to have a specific person or team who is an expert in one thing that you can work with. I don't expect that to change. Further, I think AI disruption to knowledge work will take decades to play out because it is more fundamental to the human condition (e.g. sociological/organizational) than pure intelligence.
@ericzakariasson When I’m away from my machine and I think of something I want to implement later, it would be nice to talk to an agent about the problem on mobile so I don’t forget it.
@GergelyOrosz Same experience for me in Cursor. 4.6 will often jump straight into implementation, of which it’s usually wrong, and then back track the very next prompt. Burning tokens and money. It did not use to be this way.
@cursor_ai I ran out of credits for the month and switched over to Composer 2. Here are my thoughts:
When there isn’t a predefined architecture, Composer 2 struggles a bit. But once that foundation is in place, it’s blazing fast and stays on point.
@cursor_ai I bounce between Opus and GPT, and that’s what I love about Cursor: model agnosticism. But I hate switching to GPT because the output is so verbose, while Opus is much more succinct. Is that something the Cursor harness can fix?
@PeterJThomson@cursor_ai I would prefer to choose which model to use for subagents, instead of default to Composer.
I tried setting the subagent to default to the main orchestration model but it still seems to pick Composer.