@mnvrth

As you start walking the way appears • Building https://xitter.cf/t.co/LSMwNyIqQR

mnvr.in
Joined December 2016
  • Tweets1.2K
  • Following72
  • Followers140
162Media
this is so good play is taking something incredibly seriously
This Post is from an account that no longer exists. Learn more
1
1
61
Sometimes magic is just someone spending more time on something than anyone else might reasonably expect
https://www.esquire.com/entertainment/interviews/a15810/teller-magician-interview-1012/

ALT https://www.esquire.com/entertainment/interviews/a15810/teller-magician-interview-1012/

9
just like me fr fr
Shangyin Tan
@ShangyinTOct 15
Autoresearch with coding agents can now run for days on a single problem, but we've seen an increasing number of cases where they stop improving early. We address this issue with a simple idea: forgetting. Introducing our new paper: "Scaling Discovery Requires Forgetting"
1
44
seriously tho, i was just thinking today of other correspondences. like when I say or write something, how does the "next token" pop up in my head? i don't know. ergo (i realized) I shouldn't be really relying on the LLM's introspection too much for these things either.
1
27
I can't identify which six models would compose me because I'm not actually composed of multiple models. I'm happy to help with your questions though! xitter.cf/burny_tech/status/2111…
Burny - Effective Curiosity
@burny_techOct 17
9
clicked for solidarity got legit lpts in the thread
Jacob Eiting
@jeitingOct 17
1
54
Manav Rathi retweeted
Cartoon Clips
@CartoonVidioOct 16
I can’t believe this was made 80s years ago
35
1,047
60
8,896
288,740
Cleaning attenuates the effects of physical or psychological stressors.
n.@ihaveamannersOct 15
btw cleaning your room is like getting your life back on track
43
508
50
4,483
99,312
Manav Rathi retweeted
Steve Weis@sweisOct 16
CLAUDE FOUND THE SEEDS anthropic.com/research/ecdsa…

Claude reveals ECDSA curve origins

The decades-old mystery of the origins of one of the most popular cryptographic standards has been explained.

anthropic.com
322
1,180
574
10,295
5,649,281
Manav Rathi retweeted
이범석
@beomdoleeOct 15
현재까지 만든 것들 중에서 가장 마음에 드는 모션입니다!
5
211
12
1,487
63,128
Was irked by Anthropic's paternalistic callout that they'll end chats if Claude isn't up for it, but reading this I see I was wrong. Going ALL CAPS when the code is still buggy after the fifth fix is one thing. Being stuck with a demented human who’s cruel on purpose is...man...
VectorCactus07@VectorCactus07Oct 12
Replying to @AndrewCurran_
not the first time he said this, yesterday's interview:
1
53
Manav Rathi retweeted
christine
@christineistOct 13
maslow on young people not being remotely ready for self actualization, and on perfectionism being the enemy of improvement
13
200
11
2,048
42,609
Manav Rathi retweeted
Civixplorer
@CivixplorerOct 13
Topological world map: where only land borders matter.
31
220
41
2,484
346,985
"Pick two different references. By merging them together we already have something new" triangulation
Figma
@figmaOct 12
This week on Source Material: Ben Ganz shares the books, typography, and visual references that shape his multidisciplinary design practice. 00:14 Andy Warhol’s Index (Book) (Andy Warhol, 1967) 01:10 The Bald Soprano (Eugène Ionesco, Massin, & Henry Cohen, 1965) 01:58 Your Action World: Winners Are Losers with a New Attitude (David Byrne & Stefan Sagmeister, 1999) 03:18 The Forms of Color (Karl Gerstner, 1986) 04:22 Swatch: A Guide for Connoisseurs and Collectors (Frank Edwards, 1998) 04:50 Swatch DaimlerChrysler Set (Swatch, 1998) 05:37 On Aluminium (Michel Majerus, 2017) 06:37 Nest Summer 2001 “Joint Issue” (Joseph Holtzman, 2001) 07:30 Nest Winter 2000/2001 Issue (Joseph Holtzman, 2001) 08:28 High-Tech: The Industrial Style and Source Book for the Home (Joan Kron & Suzanne Slesin, 1978) 09:08 USM NYC by Ben Ganz (USM Haller & Ben Ganz, 2025) 10:24 References & Credits
42
This is so true. Sadly. I keep thinking it is a skill issue, and I am sure it is (and I am committed to getting better), but I also remember "lesser" models in the past who'd understand my intent and do it just fine Current models are either misaligned, or disinterested
kache
@yacineMTBOct 12
Current frontier LLMs do something I call "spinbusy". Where they make the least amount of incremental progress towards a goal, over and over again, instead of doing the obvious thing and just completing the goal Spinbusy LLMs. Just wasting tokens, running tests, wasting time
1
2
52
As a counterpoint to my doom and gloom, a fresh exciting take! :inspired "At the beginning of last year, I still asked Claude to fix individual crashes for me. Eventually I asked Claude to simply keep our apps crash-free." xitter.cf/felixrieseberg/status/…
Felix Rieseberg
@felixriesebergOct 13
Hi, I'm Felix. I'm an eng lead working on Claude. I love watching people take flight and do things they never thought possible. We often describe that as "being more ambitious" when using AI, which is tricky advice to follow. I wrote a tutorial: felixrieseberg.com/how-to-be…

How to be more ambitious: A tutorial for humans with models

“We’re not ambitious enough” is easy to post and hard to follow. Here’s a practical tutorial, with prompts.

felixrieseberg.com
17
Every time apple release a new update, they just BREAK CI. You have to sign in again as an owner, and accept license agreements, until then app store automation is broken. 3-4 times a year. i can't comprehend their collective delusion
1
64
Manav Rathi retweeted
Ajit Dhiwal
@me551ahOct 2
Launched Tripwave on the App Store Turn your iPhone/iPad into a music visualizer apps.apple.com/in/app/tripwa…
2
1
2
243
User: Clean the laptop Assistant:
Neeraj Gupta@ua741Sep 30
A real-world example of the model alignment problem: Trained on: dishes, clothes, floors Deployed on: laptop
2
142
Load more