In a use case where agents can take actions or ask questions, initial tests found 3.6-flash was more likely to hedge or ask for explicit confirmation ("are you sure" or "should I proceed") after information gathering than 3-flash-preview on identical prompts/tool definitions.
We'll likely be able to prompt around this; we're just still experimenting with it. That was the first immediate thing that stood out though. For our Flash Lite tasks, 3.5 was a welcome drop in replacement with better perf.
@domjnieto27 Yep, it'll come to alpha. There's more to the redesign than just this inbox screen, so we're putting some finishing touches on the rest of it.
@donasarkar Apple Dev memberships include Xcode Cloud, so assuming you've got a repo that _should_ build fine for iOS, you can do all your building > deployment to TestFlight there (and of course, submit those builds to the App Store).
Your email is not a neutral piece of technology. It's design has potential to deeply shape your thoughts and habits.
Sunflower is built for you, to protect your attention and time.
We're often asked: why email?
Email because no one questions it anymore. It's one of the most frustrating apps on your phone. We want to show you how it can be done better.
You can expect more out of your inbox.
I’m a very patient person but if one of our engineers complained about being forced to properly align an icon at Apple I would probably not show the best side of myself.
If your HI team tells you the icon is positioned wrong, the answer is, “I’ll go figure that out.”
@domjnieto27@trysunflower There’s another navigation update in progress that brings more glass, but the primary noticeable change in this one is that the keyboard is now the correct iOS 26 keyboard, and native menus and confirmations are also using Liquid Glass.
We’re ramping up at @trysunflower and hiring another iOS engineer (Swift / UIKit / SwiftUI).
If you sweat the details and care about how software actually feels to use, we should talk.
Come build a calmer, more personal take on email.
Details below
you have to be a little obsessed and crazy to truly push on craft and design. smaller teams with passionate builders shine here.
craft is spending 7 hours on the minuscule polish thing you thought was going to take 30 minutes but you keep going deeper. the thing that is absolutely of trivial necessity that most would think is a waste of time. sane people would compromise and find another solution.
crazy people that eschew the standard way of designing and building common interactions, UI, and hardware just to see how it feels, or to tug at an inkling of a vision
then do that about 100 more times throughout your product before you're remotely proud of it. it can't be outsourced or handed off to others to implement.
Can confirm -> the new @framer logo shaders are absurdly good. Infinite customization and so much fun to play with. Dropped a few @trysunflower svgs in, and created a background sound with @floraai's new audio tools.
overcooking
you've seen this: someone ships a dashboard that shows every number with a sparkline, every action has a confirmation modal, every empty state has an animated illustration and a tagline. individually each decision made sense to someone. together it feels like chaos. nothing is in focus.
that's overcooking. not one bad decision in isolation, but the accumulation of reasonable ones that no one said no to.
AI makes this worse as the cost of adding dropped to near zero. it can build a feature, even a whole new concept in minutes. so people do. and then they do it again. the thing that started with a clear purpose slowly becomes a collection of additions that are each justifiable but collectively incoherent.
the root problem is that most "new ideas" aren't new. they're repackaging of something that already exists at a more fundamental level. a new sticker on an old concept. it feels like progress because something changed, with a new word and skin – but the thinking didn't go deeper, it just duplicated itself into confusion.
the whole has a core. you feel it once you understand the whole system. everything in it are related and balanced. when you overload it, that gravity weakens. not because any one thing is wrong – but because attention is finite and you force it everywhere.
what we need aren't more tools that make more slop. it's seeing through the chaos, and returning to what the thing actually is, and cutting everything that doesn't serve that. that's harder now, not easier. because there's always something else you could add with one more prompt.