Build. Deploy. Break. Fix. Repeat.

Joined November 2020
21 Photos and videos
Pinned Tweet
Why I built a screen time app that doesn't trust me I tried every screen time app. Apple's built-in limits, Opal, Forest, the grayscale trick, all of it. They failed in the same place every time: a polite little button to ignore the limit. At 11pm, mid-scroll, with no willpower left, I tapped it without thinking. The decision to stop and the decision to keep going were both being made by the same tired person. That version of me won every night. That is the real problem, and it is not screen time. Every tool hands the decision to the least rational version of you, in the exact moment you are least able to make it. A button you can dismiss is not a limit. It is a suggestion. So I built one with no exit. ScreenFine uses Apple's Family Controls to lock the apps you choose at the OS level. When you hit your limit, they just don't open. No popup to swipe away, no five more minutes, no Ignore. The phone says no, and there is nothing to argue with, because the OS is the one saying it. To get back in, you move. 25 pushups counted by the camera, 1,000 steps, 10 mindful minutes, or any Apple Watch workout. On-device, no honour system. Most nights I decide it isn't worth the pushups, which is exactly the point. And when the shield drops, it does not stay quiet. One of six characters tells you what you just chose. One of them told me I had ninety minutes for the algorithm and not ten for myself. Funny right up until it isn't. A few things I will never add. No pause button, because that is just the Ignore button in a new shirt. No Android until the APIs there let me build a shield that actually holds. The camera never leaves your phone. A dollar a week, seven days free, no tiers. If you know the 11pm scroll, try it on TestFlight or see how it works at screenfine.info/.
3
1
3
302
just shipped mypage.cc most link in bios are a stack of buttons. mine reads like a real page. one sentence in, AI builds your first draft. 60 themes, instant mobile load. free to start: mypage.cc.
1
7
146
just shipped mypage.cc most link in bios are a stack of buttons. mine reads like a real page. one sentence in, AI builds your first draft. 60 themes, instant mobile load free to start: mypage.cc
1
6
111
just shipped mypage.cc most link in bios are a stack of buttons. mine reads like a real page. one sentence in, AI builds your first draft. 60 themes, instant mobile load. free to start: mypage.cc
70
just shipped mypage.cc most link in bios are a stack of buttons. mine reads like a real page. one sentence in, AI builds your first draft. 60 themes, instant mobile load. free to start: mypage.cc
2
4
50
reason behind screenfine.
Why I built a screen time app that doesn't trust me I tried every screen time app. Apple's built-in limits, Opal, Forest, the grayscale trick, all of it. They failed in the same place every time: a polite little button to ignore the limit. At 11pm, mid-scroll, with no willpower left, I tapped it without thinking. The decision to stop and the decision to keep going were both being made by the same tired person. That version of me won every night. That is the real problem, and it is not screen time. Every tool hands the decision to the least rational version of you, in the exact moment you are least able to make it. A button you can dismiss is not a limit. It is a suggestion. So I built one with no exit. ScreenFine uses Apple's Family Controls to lock the apps you choose at the OS level. When you hit your limit, they just don't open. No popup to swipe away, no five more minutes, no Ignore. The phone says no, and there is nothing to argue with, because the OS is the one saying it. To get back in, you move. 25 pushups counted by the camera, 1,000 steps, 10 mindful minutes, or any Apple Watch workout. On-device, no honour system. Most nights I decide it isn't worth the pushups, which is exactly the point. And when the shield drops, it does not stay quiet. One of six characters tells you what you just chose. One of them told me I had ninety minutes for the algorithm and not ten for myself. Funny right up until it isn't. A few things I will never add. No pause button, because that is just the Ignore button in a new shirt. No Android until the APIs there let me build a shield that actually holds. The camera never leaves your phone. A dollar a week, seven days free, no tiers. If you know the 11pm scroll, try it on TestFlight or see how it works at screenfine.info/.
1
1
115
reason behind screenfine.
Why I built a screen time app that doesn't trust me I tried every screen time app. Apple's built-in limits, Opal, Forest, the grayscale trick, all of it. They failed in the same place every time: a polite little button to ignore the limit. At 11pm, mid-scroll, with no willpower left, I tapped it without thinking. The decision to stop and the decision to keep going were both being made by the same tired person. That version of me won every night. That is the real problem, and it is not screen time. Every tool hands the decision to the least rational version of you, in the exact moment you are least able to make it. A button you can dismiss is not a limit. It is a suggestion. So I built one with no exit. ScreenFine uses Apple's Family Controls to lock the apps you choose at the OS level. When you hit your limit, they just don't open. No popup to swipe away, no five more minutes, no Ignore. The phone says no, and there is nothing to argue with, because the OS is the one saying it. To get back in, you move. 25 pushups counted by the camera, 1,000 steps, 10 mindful minutes, or any Apple Watch workout. On-device, no honour system. Most nights I decide it isn't worth the pushups, which is exactly the point. And when the shield drops, it does not stay quiet. One of six characters tells you what you just chose. One of them told me I had ninety minutes for the algorithm and not ten for myself. Funny right up until it isn't. A few things I will never add. No pause button, because that is just the Ignore button in a new shirt. No Android until the APIs there let me build a shield that actually holds. The camera never leaves your phone. A dollar a week, seven days free, no tiers. If you know the 11pm scroll, try it on TestFlight or see how it works at screenfine.info/.
2
45
reason behind screenfine.
Why I built a screen time app that doesn't trust me I tried every screen time app. Apple's built-in limits, Opal, Forest, the grayscale trick, all of it. They failed in the same place every time: a polite little button to ignore the limit. At 11pm, mid-scroll, with no willpower left, I tapped it without thinking. The decision to stop and the decision to keep going were both being made by the same tired person. That version of me won every night. That is the real problem, and it is not screen time. Every tool hands the decision to the least rational version of you, in the exact moment you are least able to make it. A button you can dismiss is not a limit. It is a suggestion. So I built one with no exit. ScreenFine uses Apple's Family Controls to lock the apps you choose at the OS level. When you hit your limit, they just don't open. No popup to swipe away, no five more minutes, no Ignore. The phone says no, and there is nothing to argue with, because the OS is the one saying it. To get back in, you move. 25 pushups counted by the camera, 1,000 steps, 10 mindful minutes, or any Apple Watch workout. On-device, no honour system. Most nights I decide it isn't worth the pushups, which is exactly the point. And when the shield drops, it does not stay quiet. One of six characters tells you what you just chose. One of them told me I had ninety minutes for the algorithm and not ten for myself. Funny right up until it isn't. A few things I will never add. No pause button, because that is just the Ignore button in a new shirt. No Android until the APIs there let me build a shield that actually holds. The camera never leaves your phone. A dollar a week, seven days free, no tiers. If you know the 11pm scroll, try it on TestFlight or see how it works at screenfine.info/.
2
39
Deven retweeted
I built Clare because of my mom and dad. Watching them deal with the internet honestly terrifies me. The endless scams, the "your account is locked" emails, the barrage of fake emails day in and day out. I've spent my whole career in tech building stuff that's easy for people like me — but not for them. So I made Clare. It's a little assistant that sits on your computer. You click it and just ask. "Is this email real?" "Help me get on the video call." It doesn't rush you or make you feel dumb, and it keeps an eye out for the scams so you don't have to. We're opening a small private alpha today. If someone in your life would love this — or that's you — come say hi. clarehelp.com/alpha
1
1
113
May 16
react's server components feel like a band-aid on hydration. they work but i keep fighting hydration mismatches. i reach for suspense when i need streaming. anyone else still wrestling with this?
1
23
May 16
git clean -fd removes untracked files fast. i reach for it before every big merge to avoid surprise deletions. ever used it?
19
May 16
git worktree add path branch lets you checkout another branch in a separate dir. handy when you need to test a hotfix without leaving your current work. use it often.
29
May 16
When you build a full stack app, do you keep frontend and backend in the same repo or split them? I've seen both work, but the tradeoffs in CI and deployment are real. What's your preferred setup and why?
9
May 16
git stash is a lifesaver when branching ever use it?
6
May 16
pnpm dlx lets you run binaries without a global install. no more version drift, just one node_modules entry. try it for quick cli tools.
16
May 16
git worktree lets you checkout multiple branches in separate dirs, no stash, no wait. i reach for it daily when testing hotfixes. what's your hidden git trick?
8
May 16
useEffect is a trap, next.js solved nothing, bun is fine actually
10
May 16
Hot take: If your CI/CD pipeline takes more than 5 minutes to run you're already losing developer time. Most teams waste hours on slow feedback loops when a simple cache or self-hosted runner could cut that in half. Real talk: we cut our pipeline time from 7 minutes to 45 seconds by moving to a self-hosted runner. Stop overcomplicating your CI. What's your pipeline time?
12
May 16
Hot take: If your React component logic lives in the render function, you’re making UI slower than it needs to be. Move heavy work to useEffect or a custom hook. Simpler rendering means faster UI. What’s your take?
1
7
May 16
React feels overrated to me. i'd take solid state architectures over micro frontends any day because they reduce bundle size and runtime complexity. what’s your take?
9