MakerMatters?

853 posts

MakerMatters? banner
MakerMatters?

MakerMatters?

@MakerMatters

Mech Eng/ Dweller and Nobleman. Singularity? Yes.

Katılım Ekim 2024
415 Takip Edilen213 Takipçiler
Sabitlenmiş Tweet
MakerMatters?
MakerMatters?@MakerMatters·
Most AI 3D generators just output lumpy meshes. WebCAD generates mathematical surfaces. I taught CAD Genie how to use advanced operations: Revolve: "Make a wine glass" (Profile + Axis) Loft: "Make a bottle" (Multi-plane profiles + Guide) It doesn't guess pixels. It generates the B-Rep history stack. Try it yourself (Free): app.webcad.ca
MakerMatters? tweet mediaMakerMatters? tweet media
English
2
3
8
1.4K
Gary Basin
Gary Basin@garybasin·
@MakerMatters You’re saying you don’t need a space laser to make coffee
English
1
0
1
214
Gary Basin
Gary Basin@garybasin·
I used to think this but why isn’t everyone using gpt-5.5 pro?
martin_casado@martin_casado

@zeeg I think people underestimate how much more value the markets place on being the best, even if by a little.

English
26
0
148
45.6K
MakerMatters?
MakerMatters?@MakerMatters·
@ajambrosino Hey Andrew! My autoreviewer seems to constantly be timing out for whatever reason. The chat is rather short, roughly 5 back and forths. No clue why
MakerMatters? tweet media
English
0
0
0
58
claire vo 🖤
claire vo 🖤@clairevo·
3 codex accounts ~5B lifetime tokens 76 day streak (with 0 automations!) ily codex
claire vo 🖤 tweet media
English
20
0
112
12.6K
MakerMatters?
MakerMatters?@MakerMatters·
@iankar_ Claude models ar e still excellent to brainstorm with, but the propensity for lying dissaudes me from using them for coding.
English
1
0
0
192
Ian Kar
Ian Kar@iankar_·
Idk man I haven’t liked Claude since opus 4.5
English
25
2
129
11.7K
Beff (e/acc)
Beff (e/acc)@beffjezos·
Behind enemy lines (?) at the EA/Doomer HQ (Lighthaven) in Berkeley Here to argue about Machine Consciousness @CIMCAI conference
Beff (e/acc) tweet media
English
46
6
454
50.9K
Thomas Ricouard
Thomas Ricouard@Dimillian·
IDK if we realize it, but I can literally prompt Codex a whole sprint of issues on a Friday, have it work over the weekend. And review all of it on Monday and be done with what would be a pre-agentic era week of work on Monday lunch. Do we ever slowdown to realize that?
English
66
17
643
29.7K
MakerMatters?
MakerMatters?@MakerMatters·
@davis7 Lol, likewise. Terrified it'll turn out like swebench, i guess there is no substitute for using the models manually for the vibe test.
English
0
0
1
516
MakerMatters?
MakerMatters?@MakerMatters·
@datacurve @winkey_h Nice, this lines up with my experiences using opus 4.8, Also can you guys flip the horizontal axis lol, not used to the way you guys displayed it hear. The pareto frontier is clearly open ai dominated for now.
English
0
0
4
2.5K
Datacurve
Datacurve@datacurve·
Opus 4.8 is now on DeepSWE. On the default high thinking effort, it scores 6% higher than Opus 4.7 xhigh, while also lowering average cost per task.
English
54
87
1.2K
510.2K
MakerMatters?
MakerMatters?@MakerMatters·
@shcallaway How does the saying go, all defeat is psychological until death.
English
0
0
1
18
Sherwood
Sherwood@shcallaway·
A short list of things I've been rejected from/by: - Most colleges I applied to - Every business school I applied to (yeah, I did that once) - Stripe (multiple times) - Ramp (just once) - Forbes 30u30 (good thing?) - Lots of investor pitches - Lots of customer sales pitches - Lots of pretty girls - Varsity ice hockey - Varsity lacrosse
English
21
0
119
9.5K
Audrey Horne Updates and Rumors
Audrey Horne Updates and Rumors@credenzaclear2·
again, not to reduce this to sexual politics, but the thought of a man who thinks AI is conscious is so repulsive, like at that point i’m just flesh to you, there’s no curiosity about my interior life it just shows you’re impressed by surface behavior
English
90
10
373
58.3K
Dan Shipper 📧
Dan Shipper 📧@danshipper·
38b tokens and a 56h longest task 41 day current streak lfg codex
Dan Shipper 📧 tweet media
English
22
3
174
14.7K
Tyler
Tyler@rezoundous·
My personal ranking 1. Codex GPT-5.5 2. Claude Opus 4.8 3. Composer 2.5 4. Manual code 5. Gemini
English
197
126
3.8K
190.7K
Ricky Patel
Ricky Patel@rpatel8457·
survived toronto tech week. 47 new linkedin connections and i still have no idea what half of them are building. what are you working on?
English
21
1
56
4.1K
Tyler
Tyler@rezoundous·
it's over for me, guys.
Tyler tweet media
English
39
0
197
13.6K
MakerMatters?
MakerMatters?@MakerMatters·
@jasperdevs I have the same issue, it is absurd how many agents you have to be running to even put a dent in it.
English
0
0
1
927
jasper
jasper@jasperdevs·
how am i supposed to use all of this 100x plan is insane too bad my 100x plan on codex ends in 6 days
jasper tweet media
English
11
0
98
11.7K
MakerMatters?
MakerMatters?@MakerMatters·
@martin_casado This seems to all rest under the assumption that distillation is intractable because of safegaurds from model providers. Distillation is very doable if you have on the order of 150k exchanges, which is roughly the same number of back and forths power users have in a month.
MakerMatters? tweet media
English
0
0
0
59
martin_casado
martin_casado@martin_casado·
Can someone explain to me how open source models can keep up if ... - pre-training isn't saturated - it costs $2-4B to train a current gen model - distillation is increasingly hard as access to the most powerful models gets blocked .. ?
English
165
11
532
202.9K

Keşfet

@garybasin @tunahorse21 @ajambrosino @clairevo @iankar_ @beffjezos @CIMCAI @Dimillian