@ptrCast

https://xitter.cf/t.co/UjK3ZEETml

injuly.in
Joined January 2020
  • Tweets2.5K
  • Following279
  • Followers1.1K
213Media
Include

Only show posts containing:

Exclude

Hide posts containing:

Time range
-
Minimum likes
injuly@ptrCast4h
Replying to @octavianslash @Vexxter4sure
???? It literally is
43
injuly@ptrCastOct 18
Replying to @notnickok @DKrunks
That is not what the tool does. And this has existed for decades in animation, even in games.
25
484
injuly@ptrCastOct 16
It's insane how this book was published in 1995, and there hasn't been better material on NURBs modelling since.
Bradley Rothenberg
@brad_rothenbergOct 16
they do if they own this
4
143
injuly@ptrCastOct 10
Replying to @idlyupma
Yeah, but from the demo it seems codex's predictions are more sophisticated and intent driven
1
2
56
injuly@ptrCastOct 10
Thought experiment: If codex predicts my follow-up messages accurately, am I rly needed at that point? If labs can perfect prompt prediction, won't the models just endlessly post-train on auto-generated threads to become intelligent beyond humancomprehension?
Sharif Shameem
@sharifshameemOct 9
Codex is now accurately predicting ~1 in 4 of my user messages to it. One of the best researchers I've worked with has described their own predictions as "terrifyingly accurate". Give it a shot:
2
8
578
injuly@ptrCastOct 10
2
7
263
injuly@ptrCastOct 10
Replying to @debanwita27
Kyun
33
injuly@ptrCastOct 9
Internet censorship is getting worse (better, technically). I barely see any posts around the protest. Even the railway closure is not on my feed.
2
13
280
injuly@ptrCastOct 8
Replying to @wvo
glad to hear that!
114
injuly@ptrCastOct 8
Replying to @remilouf
I've read the Toyota Production Systems book you mentioned below -- I don't get it, wdym it transfers to software? what part?
1
929
injuly@ptrCastOct 8
Replying to @wvo
Very interesting concept, especially with the pointers. Are you planning on maintaining this? Or just an experiment?
1
2
517
injuly@ptrCastOct 6
Was not expecting this. 'Ghosts of Departed Proofs' is my all-time favorite typed FP paper. Highly recommend going through it if you haven't. If gdp-ts is as good as the tweet says, I'm moving most compile-time invariants to this.
Guillermo Rauch
@rauchgOct 5
Introducing gdp-ts: Ghosts of Departed Proofs for TypeScript. gdp-ts is a library, linter and AI skill for safer API design. Under this contract, sensitive functions require 'proofs' that the caller performed an authorization check. The typechecker verifies these proofs at compile time, preventing your team and agents from shipping catastrophic security (and other kinds of) bugs. While these patterns have existed for quite some time, especially in ecosystems like Haskell, ① human code review and ② cognitive and syntactic overhead made these solutions niche. The situation is now inverted. Agents are writing more code than we can review, and they *thrive* in tight loops with hard constraints that would frustrate us. We see this with the rise of Rust, borrow checker, code aesthetics debate and all. The README and examples model a real-world Vercel API product constraint: changing the password on a Project requires a proof of a certain role + a certain entitlement. Thanks to Matt Noonan and Ollie Charles for their research in this space. github.com/rauchg/gdp-ts
2
2
99
8,471
injuly@ptrCastOct 3
There's likely a balance that the industry (I hope) will converge on, or the agents get better with long term memory. But until then, I strongly encourage being acutely aware of your code. As for editors: I'm currently using Zed, and still looking out. Nearly quit nvim.
2
1
25
890
injuly@ptrCastOct 3
Everything I want is technically "available", but not within reach. And theres no common "language" between me and the editor. Ironically, the best editor I found—and then ditched for other reasons—was one where the author rejects all use of AI. It's called Gram. [3/n]
1
1
29
987
injuly@ptrCastOct 3
IME, Apps with unreviewed code tend to be the most impressive on the tin. Edit huge files, git integration, vim mode, LSP, plugins, semantic grep. And yet, I hit a million rough edges in my first hour of use. The interface is all over the place, hotkeys inconsistent [2/n]
2
1
36
1,067
injuly@ptrCastOct 3
Agents are can now write ~100% of most code. And yet, knowing the implementation details in-depth *is* important. Apps where devs don't care about code have a "smell". Your users can feel it. I've felt it myself while looking to switch text editors. [1/n]
Charlie Marsh
@charliermarshOct 2
"No one on my teams writes code anymore" is no longer a useful differentiator. The real question is: does anyone on your team _read_ code anymore?
5
7
2
189
9,777
injuly@ptrCastOct 3
This is really simple and clever. Makes you wonder why culling didn't occur to competitive FPS/Mobas in all these years!
Call of Duty
@CallofDutyOct 2
Few things ruin a great match faster than wondering whether an opponent earned the win or saw you through a wall. Wallhacks give cheaters information they were never meant to have. With Modern Warfare 4, we’re introducing a new way to fight back: Server Culling. Here’s how it works:
1
3
222
injuly@ptrCastOct 3
Replying to @HumanoidFrog
For most, the reason to reject AI was it's quality. This gap is on the verge of closing. Anyone who cared about the outcome rather than the process will accept AI art sooner or later.
1
93
injuly@ptrCastOct 2
Replying to @skalskip92
It's an internal model. Trained on our own data.
7
injuly@ptrCastOct 2
Replying to @skalskip92
This is the exact kind of layout and handwriting we're currently parsing. In my case, hand-filled reports from factories.
1
4
129
Load more