Alex Sexton

25.4K posts

Alex Sexton banner
Alex Sexton

Alex Sexton

@SlexAxton

Fun and cool/strong — onions are my own. Eng @pierrecomputer – prev @stripe @jquery @yayquery

Austin, TX Katılım Ocak 2008
1.2K Takip Edilen31.8K Takipçiler
Sabitlenmiş Tweet
Alex Sexton
Alex Sexton@SlexAxton·
What if instead of self-driving trucks we had direct lines between common shipping locations so traffic coulDAMMIT I invented trains again.
English
15
206
937
0
James Long
James Long@jlongster·
Remember last week when Dax asked what feature we’re shipping this week? this was it And he just publicly committed me to ship worktrees next week, lol Dax is no longer ceo or devrel, he’s a product manager (it’s good to have goals)
dax@thdxr

what pushed us to finally implement it is heavier use of worktrees now that i have sessions going across different worktrees very annoying to find them and open my editor in the right spot we'll ship the worktree feature out from the flag next week

English
7
1
84
16.3K
Christoph Nakazawa
Christoph Nakazawa@cnakazawa·
Codiff v0.5: * UX/UI/Diff Improvements * Extended the git markers the CLI understands * Happy to report the original diff that Codiff was created for has been reviewed brew install --cask nkzw-tech/tap/codiff github.com/nkzw-tech/codi…
English
2
2
26
2.6K
Ryan Jones
Ryan Jones@rjonesy·
Come build @Flighty. We need a 2nd product team. Must be elite-level. - Sr. designer - Sr. iOS + backend engineer DMs open. Include work. flighty.com/careers
Ryan Jones tweet media
English
22
23
434
85.3K
Alex Sexton
Alex Sexton@SlexAxton·
@pierrecomputer @amadeus @mdo dang, this is my first time seeing this image but im glad my height and strength are accurately represented especially as compared to mdo and amadeus. 🙏🏻
English
1
0
7
246
Alex Sexton
Alex Sexton@SlexAxton·
@damageboy @jtaby It’s just requesting a patch file directly from GitHub for the demo, everything else is client side. #codeview" target="_blank" rel="nofollow noopener">diffs.com/docs#codeview
English
1
0
1
26
damageboy
damageboy@damageboy·
@SlexAxton @jtaby Is there a guide on seeing this up for private repos, or just point an LLM and pray?
English
1
0
0
23
Alex Sexton
Alex Sexton@SlexAxton·
@birch_js @pierrecomputer that's pretty much the patch size plus the html for the syntax highlighting. always places to improve but on the big boys that's about what i'd expect. safari is the hardest for sure, at the moment.
English
1
0
1
166
Jamie Birch
Jamie Birch@birch_js·
Memory usage in Safari: - on page load: 892 MB - after scrolling the entire document: 1,470 MB That may be justified, though? Depends whether it's an actual leak or just the cost of keeping things in cache so that you can scroll back and forth. CC @pierrecomputer
Jamie Birch tweet mediaJamie Birch tweet media
English
1
0
0
788
Jamie Birch
Jamie Birch@birch_js·
I tried out DiffsHub on Safari. No idea how Twitter is gonna compress this .mov, so here goes nothing. It's pretty dang good. Some noticeable stuttering when there's both additions and deletions side-by-side. Still absolutely wipes the floor with GitHub, of course. Bravo! 👏
Pierre@pierrecomputer

diffshub[dot]com Take any public diff from GitHub and virtualize it nearly instantly, no matter how large, with DiffsHub. Built to show off our brand new CodeView component. To try it out, replace `github` with `diffshub` in your address bar.

English
3
1
112
24.6K
Alex Sexton
Alex Sexton@SlexAxton·
Ah yea, I guess what I mean is that the thing we're showing off here is our open source components. This isn't really a product we expect people to use directly (the comments aren't stateful, etc). The hope is that all of the actual players in this space use trees and diffs from us to render so that everyone can be fast in the tools they know and love. (Diffs is already in a ton of the coding apps, codex/conductor/opencode/etc) I think more folks will adopt the pattern! It's certainly possible with the primitives today!
English
2
0
1
49
Majd Taby
Majd Taby@jtaby·
@SlexAxton the problem is you need a PR on GitHub to access that state, but I need to iterate deeply with the agent before the PR exists
English
1
0
0
52
Alex Sexton
Alex Sexton@SlexAxton·
@dalmaer my honest guess would be if promised to never let anyone use this (the opposite)
English
1
0
0
176
patagucci perf papi
patagucci perf papi@kenwheeler·
the world needs an italian american nj based super hero that defeats enemies (fuggin jerkawfs) with a pizza peel
patagucci perf papi tweet media
English
25
1
102
3.4K
Igor bedesqui
Igor bedesqui@bedesqui·
hey @pierrecomputer , i just found this from november last year. just saying, i think i independently arrived at diffs before knowing yall were working on it!!!!!
Igor bedesqui tweet media
English
1
0
1
361
Joe
Joe@jrchsn·
@SlexAxton @fat @amadeus @necolas @mdo i appreciate that, but my fault tbh - speakers have to be on full, how else will claude notify me that i am needed back at my desk
English
1
0
1
31
David Flanagan
David Flanagan@rawkode·
@mitchellh Nonsense. The problem with code review isn’t speed, it’s comprehension and this doesn’t increase that at all. AI makes PRs bigger, and you’re smart enough to understand that reading a large PR quicker doesn’t solve anything.
English
5
0
12
3.5K
Mitchell Hashimoto
Mitchell Hashimoto@mitchellh·
This is how performant PR review could be. On any forge. Pierre is showing us that the only thing holding that back is a skill issue. Excellent ship here! They’re on fire!
Pierre@pierrecomputer

diffshub[dot]com Take any public diff from GitHub and virtualize it nearly instantly, no matter how large, with DiffsHub. Built to show off our brand new CodeView component. To try it out, replace `github` with `diffshub` in your address bar.

English
36
122
3K
251.7K
Alex Sexton
Alex Sexton@SlexAxton·
we agree. i think the goal of this tech demo (fully just a demo, doesn't store state, etc) is to show off the open source components we offer and show that they won't be a bottleneck for whatever you're cooking. i certainly don't have any interest in reviewing anything this size, but also have hit many PRs in github that I *did* want to review that it couldn't render. The goal was to show the gap.
English
1
0
3
84
Danni Friedland
Danni Friedland@DanniFriedland·
@mitchellh this is super nice as a tech demo. but no one reviewing that amount of code.. not in a scrolling manner. hopefully not at all.
English
1
0
7
1.3K
Alex Sexton
Alex Sexton@SlexAxton·
@jrchsn @fat @amadeus @necolas @mdo the lawyers asked me to note that any damage to speakers or ear drums is the sole responsibility of the end user and pierre makes no claims to the decibels therein.
English
1
0
0
41