Jad Ghalayini (@technicallyrain) header
Jad Ghalayini (@technicallyrain) profile picture

Jad Ghalayini

@technicallyrain

(locally) nameless one @autoresearchinc (YC26)

tekne.dev
Joined February 2020
  • Tweets60
  • Following192
  • Followers347
18Media
Photo by Jad Ghalayini (@technicallyrain)Photo by Jad Ghalayini (@technicallyrain)Photo by Jad Ghalayini (@technicallyrain)Photo by Jad Ghalayini (@technicallyrain)Photo by Jad Ghalayini (@technicallyrain)Photo by Jad Ghalayini (@technicallyrain)Photo by Jad Ghalayini (@technicallyrain)Photo by Jad Ghalayini (@technicallyrain)
It is not current year, but $CURRENT_YEAR
2
110
Who would have thought that otoro can elevate even a california roll
Image from Jad Ghalayini (@technicallyrain)
3
81
There lies his crown in waters deep, till Waleed finally gets his PhD (and joins @autoresearchinc)
Image from Jad Ghalayini (@technicallyrain)
3
4
219
Kraken b like
Image from Jad Ghalayini (@technicallyrain)
2
98
Replying to @thomasboser
@thomasboser
Image from Jad Ghalayini (@technicallyrain)
1
31
Replying to @lambdabetaeta
Since I'm formalizing corridors and rooms in an institution, this is the natural next step really...
15
Replying to @lambdabetaeta @deepmatmul
Fixed, thanks: github.com/hiloopai/silmaril…
Prove exact PCF representation of partial recursive functions by imbrem · Pull Request #18 ·...

Prove exact PCF representation of partial recursive functions by imbrem · Pull Request #18 ·...

The existing PCF representation theorem guarantees the outputs of a partial recursive function when they are defined. It leaves undefined inputs unconstrained. With call-by-name composition, a cons...

github.com
3
34
Back in SF, time to unleash the kraken
Image from Jad Ghalayini (@technicallyrain)
9
238
Replying to @deepmatmul
Putting my bet out there: it totally will
1
134
Gift for the plane ride to SF from the evil Dr. Proselkov. Review: thank *god* for extroversion
Image from Jad Ghalayini (@technicallyrain)
3
168
Replying to @TaliaRinger
watch this space
2
79
Replying to @_vatsadev
a king among interns
1
1
28
Jad Ghalayini retweeted
0xSero (@0xSero) profile picture
0xSero
@0xSeroNov 3
Formal verification is the only way, fully convinced. Shoutout to Taelin for seeing it before most of us and building it out. Someone needs to do a YouTube series “formal verification for agent acceleration”
lauren (@poteto) profile picturelauren
@potetoNov 3
maybe a hot take but i now consider adversarial reviews mostly a waste of tokens. far better to use those tokens on empirical combined with formal verification. i have updated pstack to use the interrogate skill less, and eventually i think i may remove it altogether i have found very little value with code review in the abstract, agents tend to just get lost in the nitpicks and yagnis
37
24
3
914
30,987
Replying to @deepmatmul
we keep it compositional, yo
1
2
141
Need to get my photo crossed out…
Image from Jad Ghalayini (@technicallyrain)
2
5
316
Put a heart ❤️ next to a coding problem, and Claude & GPT write recursive code 100% of the time (0% baseline). They also can't tell you why! Excited to share “A Heart Means Recursion” from my summer at @GoogleDeepMind, on how VLMs read & communicate with visual symbols. 🧵
32
91
36
970
72,912
Replying to @thomasboser
*in WASM
3
23
Jad Ghalayini retweeted
Thomas Boser (@thomasboser) profile picture
Thomas Boser
@thomasboserNov 4
bro what if i made a proof checker to check your proof checker to check your proof checker to check your proof checker to check your proof checker to check your proof checker to check your proof checker to check your proof checker to check your proof checker
Karan Brar (@deepmatmul) profile pictureKaran Brar
@deepmatmulNov 4
Replying to @deepmatmul
We also found bugs along the way. hol.mm, the HOL database in the set.mm repo, proves T ⊢ F: its polymorphic constants don't carry their type. Putting the Wasm spec's SpecTec sources into Lean turned up errata in 1.0, 2.0 and 3.0. We're in the process of reporting these bugs.
1
3
10
597
Load more