Paul Jump

717 posts

Paul Jump banner
Paul Jump

Paul Jump

@paulljump

I once told Buzz Aldrin that I liked his watch.

New York, NY Katılım Aralık 2008
286 Takip Edilen223 Takipçiler
Paul Jump
Paul Jump@paulljump·
Still have my Claude pro account so will through some jobs that way each session block and 8% of the time when I want to use Fable it switch. That experience is so terrible. You can tell they don’t want to build this as a consumer product.
English
0
0
2
84
Josh Cohenzadeh
Josh Cohenzadeh@jshchnz·
Serious question not meant in a snarky way: Are there any verifiable examples of fully vibecoded software reliably generating $1M+ in ARR?
English
108
0
183
65.3K
David Scott Patterson
David Scott Patterson@davidpattersonx·
GPT-5.6 gets an initial IQ score of 136, which is smarter than 99% of humans. GPT-5.6 is the first model to score over 130.
David Scott Patterson tweet media
English
113
219
2.1K
251.7K
Yiatin Chu
Yiatin Chu@ycinnewyork·
23% of households on the UES and UWS own a car. The City is "retaking" nearly 30,000 parking spots in these neighborhoods by 2032 for trash bins. So much for making the city more affordable for families. Another one added to the list of why families leave NYC. gothamist.com/news/parts-of-…
English
90
53
220
86.1K
Paul Jump
Paul Jump@paulljump·
@aarmlovi I think the meta point is that maybe this is the actual reason new housing isn’t being built
English
0
0
2
128
Alex Armlovich
Alex Armlovich@aarmlovi·
Metro NYC's ~20M people would have to build Tokyo-scale transit before the region approaches Tokyo's ~40M pop. But if we relaxed residential growth control laws within 0.5mi of rail stations, metro NYC could easily add 2M to 5M homes *on existing rail*. Enough for 30M people
Alex Armlovich tweet mediaAlex Armlovich tweet media
Alex Armlovich@aarmlovi

Tokyo and NYC were roughly the same size when NYC passed the 1961 downzoning Today NYC is still under 20M people, while Tokyo is almost 40M people NYC looks big but it's actually quite small for a country the size of America

English
5
13
154
11.8K
Dillon Mulroy
Dillon Mulroy@dillon_mulroy·
nobody actually has things figured out btw
Dillon Mulroy tweet media
English
14
16
312
19.2K
Paul Jump
Paul Jump@paulljump·
@ramez This makes me feel like they think we couldn’t do any of this before they showed up
English
0
0
0
6
Rand
Rand@rand_longevity·
@thsottiaux the best week in ai history is happening rn
English
4
1
23
1.4K
Tibo
Tibo@thsottiaux·
Hello. We have reached 8M active users across Codex and ChatGPT Work. We are once again resetting the usage limits for all. And we continue to not have the 5h rate limit as well, allowing everyone to explore the boundaries of GPT-5.6 Sol and discover how ambitious you can be. See you tomorrow for more updates on our growth!
Sam Altman@sama

5.6 sol growth is insane. the inference team has done heroic work to be able to support demand. we are going to move mountains to continue to scale, but it is possible there are some hiccups soon.

English
2K
964
16.6K
2.4M
Paul Jump
Paul Jump@paulljump·
The question is capex + time vs opex + speed.
English
1
0
0
56
Paul Jump
Paul Jump@paulljump·
I am building simulated companies right now and it’s getting very expensive. I’ll share a longer form write up on this soon, but here are some basic economics 🧵
English
1
0
0
252
Matt Shumer
Matt Shumer@mattshumer_·
Is it just me or is Claude Fable eating usage limits way faster today?
English
141
2
359
58.9K
Paul Jump
Paul Jump@paulljump·
Enough signal to continue. Next: 308 sources selected by mapping the work a firm must perform, finding the official materials capable of teaching each part, and freezing the it before mining. Will keep this updated.
English
0
0
0
20
Paul Jump
Paul Jump@paulljump·
A few recognizable pieces of professional work survived: classifying creditors, determining which forms a client’s facts trigger, checking signatures and disclosures, and assembling parts of a filing package.
English
1
0
0
26
Dave Gerhardt
Dave Gerhardt@davegerhardt·
My company is 8 people. Profitable. No outside capital. $5M revenue this year (doubled every year since 2023). I can work from home. Do school dropoff/pickup. Have a snack with my kids after school. But everyone asks how my little lifestyle business is going? So if you aren't sleeping on the office floor you're not a real entrepreneur? SMH
English
144
23
1.7K
205.1K
Paul Jump
Paul Jump@paulljump·
@MaaSonder I’d actually sign on to this. I actually don’t think it was intentional but rather they just made a lot of assumptions which for a company with a safer brand is damming
English
0
0
0
17
Qwinah
Qwinah@MaaSonder·
Anthropic is becoming a joke at this point Every week something changes, temporary limit increases, weekly credits, separate Sonnet and Opus caps, Fable credits, 5-hour limits, and rolling weekly restrictions It shifts constantly, forcing us to check dashboards daily and worry about hitting walls early This has turned a paid service into a nightmare. Yes compute is expensive, but that’s their problem to solve, not something to offload onto customers through confusing credits and ever-changing limits Paying users should be able to relax and work, not feel like addicts rationing their next dose Anthropic needs to rethink this entire system from a customer-first perspective Fix it now, before the industry normalizes constant quota anxiety instead of real productivity. We just want to pay and use the tool, not struggle with it
Qwinah tweet media
Qwinah@MaaSonder

Anthropic extended Fable 5 again, and at this point i don’t even think they know what to do If compute capacity was truly the issue, they would have enforced the original deadline. Instead we keep getting one week extensions, which tells me the backlash was bigger than expected GPT-5.6 and Grok 4.5 have both launched, and the last thing Anthropic wants is a wave of subscribers saying “they just took away their best model.” So they’re stalling. Not because they suddenly found extra GPUs, but because they don’t have a better answer yet At this point the repeated extensions are hurting credibility more than the original decision itself

English
21
22
259
20.2K