hi i'm blake. i use this account to break down interfaces that should work but don't, and share my honest takes on AI and what it's changing about how we build.
things i post about:
β’ AI tools i've used for more than a week
β’ UX decisions that make me want to open the inspector
β’ things that shipped and worked, because those exist too
if any of that sounds useful, stick around
stripe ran a 50 million line ruby codebase migration with fable 5 in a day
a task that would've taken a full team two months.
i keep waiting for one of these case studies to not hold up. still waiting
@Dexerto a "server offline" screen as a farewell message is a bold choice. zero context for anyone who hasn't been following the news, looks identical to an actual outage. someone on the team definitely argued against this
@Pirat_Nation "we built a universe" is a nice send-off message but shipping a "server offline" screen that looks identical to an actual outage on your final update day is a pretty avoidable UX failure
@Pirat_Nation "our business is not healthy and we have a plan but can't talk about it yet" is not a strategy. that's what you say when you don't have one
apple shipped a personalization slider for liquid glass today. you can go from ultra-clear to fully tinted
giving users control over a design system element instead of just picking for them is the right call
most apps still don't do this and it's annoying
@PolymarketMoney siri ai not launching in the eu because of the DMA, not launching in china because of regulators. the "global" launch is getting smaller by the region
@InternetH0F EU users getting a stripped down siri because of "compliance" is a choice apple made. they have the resources to figure it out. they just haven't decided EU users are worth the effort yet
@Pirat_Nation a game about psychological horror that never showed gameplay in 6 years and got cancelled as "just a concept" the real horror was the project management
@MKBHD the realtime preview before the generative fill is the right call. you're not guessing what you'll get, you're approving it first. more AI tools should work this way
you can go from nothing to a working interface in an afternoon now
but you still can't tell from looking at the output whether it got what you meant or just produced something that passes at a glance
nobody building these tools seems particularly bothered by that either
800 fixes in a month that would've taken a human 4 years. cleanup, edge cases, bugs in someone else's code that nobody wants to touch. the stuff that just sits there forever because it's never worth the time. that's what actually changes here. not just the hard work getting automated. the deprioritized work finally getting done
@CultureCrave a 20 year old filmmaker whose entire career started on the internet just called a trade report "hallucinated" he knows exactly what word he picked
the store page literally said marathon was included. people bought it based on what the listing said. and instead of just honoring it a $40 game, for a handful of players they quietly edited the description and told people to go fight playstation support for a refund. marathon is already fighting for players. this is not the hill to die on
so microsoft just launched Scout. an always-on agent that lives in your calendar and inbox like an actual employee
you name it, train it over time, and the more you invest in it, the harder it is to leave
they know exactly what they're doing