Corbin Crutchley
@crutchcorn.dev
VP of Software Engineering | @GitHub.com Star | Microsoft MVP | @playfulprogramming.com Partner | @TanStack.com Maintainer | Twitch Streamer (http://twitch.tv/crutchcorn) 💅💅💅
Every day is learning day in Zeu-land 🌈
February 9, 2026 at 5:24 PM
Every day is learning day in Zeu-land 🌈
Also featured in that video:
This weird ass expression I made
Thanks for sharing THAT @jason.energy - is this retribution for all the bad jokes I made that weekend? 😂
This weird ass expression I made
Thanks for sharing THAT @jason.energy - is this retribution for all the bad jokes I made that weekend? 😂
February 9, 2026 at 5:23 PM
Also featured in that video:
This weird ass expression I made
Thanks for sharing THAT @jason.energy - is this retribution for all the bad jokes I made that weekend? 😂
This weird ass expression I made
Thanks for sharing THAT @jason.energy - is this retribution for all the bad jokes I made that weekend? 😂
Welp... This is annoying
I used some tricks to extract the NPU + Win drivers from a @frame.work, then Copilot + Opus to reverse engineer the ability to apply those drivers to my GPD Win Max 2..
But it doesn't have the features I thought it had (background replacement), so I did all of this for not
I used some tricks to extract the NPU + Win drivers from a @frame.work, then Copilot + Opus to reverse engineer the ability to apply those drivers to my GPD Win Max 2..
But it doesn't have the features I thought it had (background replacement), so I did all of this for not
February 8, 2026 at 7:11 AM
Welp... This is annoying
I used some tricks to extract the NPU + Win drivers from a @frame.work, then Copilot + Opus to reverse engineer the ability to apply those drivers to my GPD Win Max 2..
But it doesn't have the features I thought it had (background replacement), so I did all of this for not
I used some tricks to extract the NPU + Win drivers from a @frame.work, then Copilot + Opus to reverse engineer the ability to apply those drivers to my GPD Win Max 2..
But it doesn't have the features I thought it had (background replacement), so I did all of this for not
Excited to be part of the GitHub Stars program for 2026 again! 🚀
February 3, 2026 at 6:58 PM
Excited to be part of the GitHub Stars program for 2026 again! 🚀
Just hit 1,000 followers on GitHub!
Check it out for yourself:
github.com/crutchcorn/
How exciting, thank you for joining me for my developer journey!
Check it out for yourself:
github.com/crutchcorn/
How exciting, thank you for joining me for my developer journey!
January 30, 2026 at 6:34 AM
Just hit 1,000 followers on GitHub!
Check it out for yourself:
github.com/crutchcorn/
How exciting, thank you for joining me for my developer journey!
Check it out for yourself:
github.com/crutchcorn/
How exciting, thank you for joining me for my developer journey!
tauri, what is this??
January 25, 2026 at 4:19 AM
tauri, what is this??
OK so to be faiiiir: Boox has a (problematic) SDK that allows you to integrate with what I assume is the same underlying API.
I'll loop back to why it's problematic in the next posts, but see this video for why it doesn't even fix all of OneNote:
I'll loop back to why it's problematic in the next posts, but see this video for why it doesn't even fix all of OneNote:
January 2, 2026 at 9:24 AM
OK so to be faiiiir: Boox has a (problematic) SDK that allows you to integrate with what I assume is the same underlying API.
I'll loop back to why it's problematic in the next posts, but see this video for why it doesn't even fix all of OneNote:
I'll loop back to why it's problematic in the next posts, but see this video for why it doesn't even fix all of OneNote:
....
It's because in Boox's firmware (as of 3.2.2) they hardcode OneNote, WPS, and EverNote to use their inking stuff rather than the built-in stuff
...
Dead end. 😭
It's because in Boox's firmware (as of 3.2.2) they hardcode OneNote, WPS, and EverNote to use their inking stuff rather than the built-in stuff
...
Dead end. 😭
January 2, 2026 at 9:07 AM
....
It's because in Boox's firmware (as of 3.2.2) they hardcode OneNote, WPS, and EverNote to use their inking stuff rather than the built-in stuff
...
Dead end. 😭
It's because in Boox's firmware (as of 3.2.2) they hardcode OneNote, WPS, and EverNote to use their inking stuff rather than the built-in stuff
...
Dead end. 😭
Giving GameActivity a shot to see if the bottleneck is JNI input throughput
January 2, 2026 at 7:55 AM
Giving GameActivity a shot to see if the bottleneck is JNI input throughput
Apparently 0.3ms still isn't fast enough...
And I don't think eInk needs 3000fps so maybe time to try a different approach...
And I don't think eInk needs 3000fps so maybe time to try a different approach...
January 2, 2026 at 7:35 AM
Apparently 0.3ms still isn't fast enough...
And I don't think eInk needs 3000fps so maybe time to try a different approach...
And I don't think eInk needs 3000fps so maybe time to try a different approach...
Bought an eInk Android tablet (Boox) for a friend around Christmas.
While developing an app for it, we came to realize that only first-party drawing apps (using their C library) have acceptable latency.
This is what a stripped down Jetpack Ink API looks like on it:
While developing an app for it, we came to realize that only first-party drawing apps (using their C library) have acceptable latency.
This is what a stripped down Jetpack Ink API looks like on it:
January 2, 2026 at 5:33 AM
Bought an eInk Android tablet (Boox) for a friend around Christmas.
While developing an app for it, we came to realize that only first-party drawing apps (using their C library) have acceptable latency.
This is what a stripped down Jetpack Ink API looks like on it:
While developing an app for it, we came to realize that only first-party drawing apps (using their C library) have acceptable latency.
This is what a stripped down Jetpack Ink API looks like on it:
I've been wanting an e-ink tablet to help me focus while doing writing...
But then keep realizing all of the problems that would arise (window management, code editing, et al)
So instead I wrote a shader for my Windows desktop that mimics the style.
Looks shockingly convincing to my eyes
But then keep realizing all of the problems that would arise (window management, code editing, et al)
So instead I wrote a shader for my Windows desktop that mimics the style.
Looks shockingly convincing to my eyes
December 13, 2025 at 7:17 AM
I've been wanting an e-ink tablet to help me focus while doing writing...
But then keep realizing all of the problems that would arise (window management, code editing, et al)
So instead I wrote a shader for my Windows desktop that mimics the style.
Looks shockingly convincing to my eyes
But then keep realizing all of the problems that would arise (window management, code editing, et al)
So instead I wrote a shader for my Windows desktop that mimics the style.
Looks shockingly convincing to my eyes
🧵 6/6 This information is a huge boon to writing complex operations that need to operate through a loop at inconsistent rates; like a parser or lexer.
Hope you've been able to learn something from this thread!
Hope you've been able to learn something from this thread!
December 2, 2025 at 8:52 PM
🧵 6/6 This information is a huge boon to writing complex operations that need to operate through a loop at inconsistent rates; like a parser or lexer.
Hope you've been able to learn something from this thread!
Hope you've been able to learn something from this thread!
🧵 5/6 And because of that, we're even able to modify the state of `i` even if we do have the `i++`.
The following code is the same as writing `i+=2` instead of `i++`:
The following code is the same as writing `i+=2` instead of `i++`:
December 2, 2025 at 8:52 PM
🧵 5/6 And because of that, we're even able to modify the state of `i` even if we do have the `i++`.
The following code is the same as writing `i+=2` instead of `i++`:
The following code is the same as writing `i+=2` instead of `i++`:
🧵 4/6 But that's not all, we can even remove the `i++` and do it ourselves in the body of the `for` loop:
December 2, 2025 at 8:52 PM
🧵 4/6 But that's not all, we can even remove the `i++` and do it ourselves in the body of the `for` loop:
🧵 3/6 This means that our simple `for` loop above can be written using `while(true)`, like so:
December 2, 2025 at 8:52 PM
🧵 3/6 This means that our simple `for` loop above can be written using `while(true)`, like so:
🧵 2/6 If we use an IIFE:
(() => 1)()
(This one immediately runs the function and returns `1`)
To `console.log` for each item in the `for` loop parens, we can see that:
`let i = 0` runs once
`i < 3` runs before each iteration
`i++` runs after each iteration
(() => 1)()
(This one immediately runs the function and returns `1`)
To `console.log` for each item in the `for` loop parens, we can see that:
`let i = 0` runs once
`i < 3` runs before each iteration
`i++` runs after each iteration
December 2, 2025 at 8:52 PM
🧵 2/6 If we use an IIFE:
(() => 1)()
(This one immediately runs the function and returns `1`)
To `console.log` for each item in the `for` loop parens, we can see that:
`let i = 0` runs once
`i < 3` runs before each iteration
`i++` runs after each iteration
(() => 1)()
(This one immediately runs the function and returns `1`)
To `console.log` for each item in the `for` loop parens, we can see that:
`let i = 0` runs once
`i < 3` runs before each iteration
`i++` runs after each iteration
🧵 1/6 `for` loops in JavaScript are more complex than you think they are.
Let's start with a basic `for (let i = 0;)`...
Here we can see it counting from 0 - 2.
But everything in the parens for the `for` loop aren't executed all-at-once.
Let's start with a basic `for (let i = 0;)`...
Here we can see it counting from 0 - 2.
But everything in the parens for the `for` loop aren't executed all-at-once.
December 2, 2025 at 8:52 PM
🧵 1/6 `for` loops in JavaScript are more complex than you think they are.
Let's start with a basic `for (let i = 0;)`...
Here we can see it counting from 0 - 2.
But everything in the parens for the `for` loop aren't executed all-at-once.
Let's start with a basic `for (let i = 0;)`...
Here we can see it counting from 0 - 2.
But everything in the parens for the `for` loop aren't executed all-at-once.
I was just on the season 2 finale of @codetv.dev!
It's not out yet, but had an absolute blast with @jason.energy, my code partner @zeu.dev, and the rest of the crew!
Beyond excited for when the episode goes live - I'll share more then!
It's not out yet, but had an absolute blast with @jason.energy, my code partner @zeu.dev, and the rest of the crew!
Beyond excited for when the episode goes live - I'll share more then!
November 17, 2025 at 11:05 PM
I was just on the season 2 finale of @codetv.dev!
It's not out yet, but had an absolute blast with @jason.energy, my code partner @zeu.dev, and the rest of the crew!
Beyond excited for when the episode goes live - I'll share more then!
It's not out yet, but had an absolute blast with @jason.energy, my code partner @zeu.dev, and the rest of the crew!
Beyond excited for when the episode goes live - I'll share more then!
I have never laughed this hard in my life.
October 29, 2025 at 7:01 AM
I have never laughed this hard in my life.
Most companies have similar business needs for their software, but not all needs are met equally
IE: Error handling. Critical for biz needs. Yet so many companies I've seen absolutely neglect it. Why? Not enough engineers focusing on that discipline
Get good at that stuff and you'll be unstoppable
IE: Error handling. Critical for biz needs. Yet so many companies I've seen absolutely neglect it. Why? Not enough engineers focusing on that discipline
Get good at that stuff and you'll be unstoppable
October 26, 2025 at 3:02 PM
Most companies have similar business needs for their software, but not all needs are met equally
IE: Error handling. Critical for biz needs. Yet so many companies I've seen absolutely neglect it. Why? Not enough engineers focusing on that discipline
Get good at that stuff and you'll be unstoppable
IE: Error handling. Critical for biz needs. Yet so many companies I've seen absolutely neglect it. Why? Not enough engineers focusing on that discipline
Get good at that stuff and you'll be unstoppable
October 22, 2025 at 9:21 AM
Coming soon to a @tanstack.com near you
October 8, 2025 at 2:30 AM
Coming soon to a @tanstack.com near you
Turns out that the Activity API uses `display: none !important` to hide UI elements.
Super cool API; very excited for it
Super cool API; very excited for it
October 7, 2025 at 6:13 PM
Turns out that the Activity API uses `display: none !important` to hide UI elements.
Super cool API; very excited for it
Super cool API; very excited for it