Scotty Weeks

9.5K posts

Scotty Weeks

Scotty Weeks

@scottyweeks

Fiction and bad ideas, mostly.

New York Katılım Eylül 2009
1K Takip Edilen1.3K Takipçiler
Scotty Weeks retweetledi
Soumitra Shukla
Soumitra Shukla@soumitrashukla9·
This chart is very related to the Substack @alexolegimas and I wrote yesterday. One explanation for this pattern is that this is basically the O-ring "focus effect" playing out in real time for software engineering. Engineering is a high-dimensional job comprising coding, testing, and debugging, but the job has enough complementary dimensions that the worker becomes more productive rather than displaced. And if demand for software is elastic (which it appears to be, given that cheaper/faster development means more products get built), then the model predicts exactly what the chart shows: more hiring, not less. A counterexample here would be low-dimensional tech roles showing the opposite trend over the same time period.
Lenny Rachitsky@lennysan

Engineering job openings are at the highest levels we’ve seen in over 3 years There are over 67,000 (!!!) eng openings at tech companies globally right now, with 26,000 just in the U.S. We don’t know if there would have been more open roles if not for AI or if AI is actually leading to more open roles, but since the start of this year, the increase in open eng roles is accelerating even more.

English
10
21
89
16.3K
Scotty Weeks retweetledi
4nzn
4nzn@paoloanzn·
the CEO of Vercel saying the saas apocalypse is real because they replaced internal tools with AI-generated apps is so funny to me… you run a software company bro. you have engineers everywhere. if YOU couldn't replace your own internal tooling with vibe-coded apps that would be embarrassing honestly, AI or not that's like a mechanic saying cars are easy to fix and concluding nobody needs mechanics anymore. yeah no shit it's easy for YOU 90% of businesses out there don't have a single person on staff who knows what an API call even is. they're not replacing their CRM with something they prompted in claude code the disconnect is wild. these tech CEOs live in such a bubble that they genuinely think their experience is universal. your company literally builds deployment infrastructure, obviously you can ship internal tools fast the SaaS apocalypse might come eventually idk, but pls get back in touch with reality
Guillermo Rauch@rauchg

Almost every SaaS app inside Vercel has now been replaced with a generated app or agent interface, deployed on Vercel. Support, sales, marketing, PM, HR, dataviz, even design and video workflows. It’s shocking. The SaaSpocalypse is both understated and overstated. Over because the key systems of record and storage are still there (Salesforce, Snowflake, etc.) Understated because the software we are generating is more beautiful, personalized, and crucially, fits our business problems better. We struggled for years to represent the health of a Vercel customer properly inside Salesforce. Too much data (trillions of consumption data points), the ontology of Vercel was a mismatch to the built-in assumptions, and the resulting UI was bizarre. We generated what we needed instead. When you don’t need a UI, you just ask an agent with natural language. We’ve also been moving off legacy systems with poor, slow, outdated, and inconsistent APIs, as well as just dropping abstraction down to more traditional databases. UI is a function 𝑓 of data (always has been), and that 𝑓 is increasingly becoming the LLM.

English
84
116
2.4K
235.6K
Scotty Weeks
Scotty Weeks@scottyweeks·
I mean I get the arguments that juniors aren’t being hired but, we’re also in a weird environment. Legacy tech is crashing and AI enabled tech is booming. That means more seniors until the dust settles and you can actually onboard a junior in a meaningful way
English
0
0
0
6
i/o
i/o@avidseries·
Liberal academic @Musa_alGharbi's observations when he moved to Manhattan: "One of the first things that stood out to me is that there’s a racialized caste system here that everyone takes for granted. You have disposable servants who will clean your house, watch your kids, walk your dogs, deliver food to you... mostly minorities and immigrants and disproportionately women... And this is basically taken for granted in New York, that this is the way society operates. And yet... this is not how things are in many other parts of the country. Most other places, the person buying a pair of shoes and the person selling them are likely to be the same race — white — and the gaps between the buyer and the seller are likely to be much smaller. Even the most sexist or bigoted rich white person in many other contexts wouldn’t be able to exploit women and minorities the same way as the typical liberal professional in a city like Seattle or New York; the infrastructure simply isn’t there. It’s these progressive bastions associated with the knowledge economy that have these well-oiled machines for casually exploiting the vulnerable, desperate and disadvantaged. And it’s largely Democratic-voting professionals who take advantage of them. A few months after I arrived at Columbia, Trump won. I expected this to happen, but for most people, that was not the expectation. So here at Columbia, the day after Trump won, a lot of the students claimed to be so traumatized that they couldn’t do tests or homework. They needed time off. Now there are two things striking about that to me. First, these are students at an Ivy League school, overwhelmingly people from wealthy backgrounds — and even if they don’t come from wealth, they’re likely to be well-positioned... [but these] students seemed to view themselves as somehow uniquely vulnerable to Trump and his regime, as being especially threatened or victimized. And so they demanded all of these accommodations for themselves. Meanwhile, there was this whole other constellation of people [mostly minorities and immigrants] around them who seemed to be literally invisible to them. The people doing all the work on the campus... these ignored laborers — the people with the most at stake in this election — [were not] saying they needed time off because they were too traumatized. They showed up to work the next day and did their jobs. They weren’t making a scene, sobbing as they scrubbed rich kids’ mess out of the toilets. The juxtaposition was sobering... When I left campus, walking around the Upper West Side, or other affluent parts of Manhattan, similar scenes were playing out. Nor was New York City unique in this regard. Other knowledge economy hubs had similar scenes playing out. And the same drama that was playing out in Columbia was unfolding at colleges and universities across the country. This is precisely what I found so troubling, so difficult to shake off: It wasn’t about my own school. It was about this broader disjuncture between knowledge-economy elites, their narratives about the world, and the realities on the ground."
English
78
444
2.4K
121.6K
Scotty Weeks retweetledi
Charles Murray
Charles Murray@charlesmurray·
One of the more charming aspects of the American mindset has been that so many people who objectively are working class or upper class answer poll questions about their class as "middle." They aren't necessarily lying or in error. Many are saying something important and positive about how they see themselves relative to other Americans.
Musa al-Gharbi@Musa_alGharbi

If there is a genre I truly hate in the @nytimes it's the frequent pieces featuring folks with multi-millions in assets and multiple hundreds of thousands of dollars in yearly household income describing themselves as middle class, and being credulously understood that way, while professionals with healthy six figure incomes engage in poverty larping because, of course, if these other people are "middle class" then they must themselves be "poor" or "working class" or "just getting by." More here: musaalgharbi.com/2025/01/01/how…

English
36
20
291
48.7K
Scotty Weeks retweetledi
Nathan Witkin
Nathan Witkin@NateWitkin·
Context rot, insufficient memory, poor long term planning, compounding hallucination risk, drift toward middle of training distribution (away from the juicy, novel ideas at the tails). In general, complex, longer-term tasks require hard constraints. You impose a plan, and follow it. You check your work at stage 1, so that any errors aren't transmitted to stage 2. LLMs, since they're indeterministic, cannot set such constraints, at least not ones that work without fail. The more such constraints a task requires, the less likely an LLM will complete it autonomously, since each one represents a 'link' in a chain with x probability of failure. >"But we've already seen LLMs succeed at incredibly hard math!" Yeah, after a gazillion trials with a ton of human hand-holding. Absent expert guidance and the budget for an arbitrarily large number of trials, they will make a costly mistake at some point. And yes, they will get better at doing their own filtering, but the issue is the same problems arise at the next layer of abstraction (i.e. for LLM 2 checking the work of LLM 1). For these and other reasons, I'm skeptical of the idea of autonomous AI research, at least if I'm meant to read "autonomous" literally. Very hard for me to see how you could cut humans fully out of the loop.
English
1
5
46
4.3K
Scotty Weeks retweetledi
Séb Krier
Séb Krier@sebkrier·
Hard to find people who simultaneously 'get' language models deeply without getting lost in them, have a good understanding of human behaviours, don't approach every sociopolitical question as binaries, understand systemic dynamics and complexity without robbing individuals of agency, hold opposing ideas in mind while retaining the ability to function, and can both zoom in and out of all the above without overindexing on any particular aspect.
English
49
24
378
21.2K
Scotty Weeks retweetledi
Noah Smith 🐇🇺🇸🇺🇦🇹🇼
Noah Smith 🐇🇺🇸🇺🇦🇹🇼@Noahpinion·
The transformation of liberals into "progressives" since 2013 has seriously been the worst thing to happen to American culture in my lifetime. Just absolutely gutted so much of what was good about this country. End of a golden age.
English
311
720
8.7K
1.1M
Scotty Weeks retweetledi
𝖓𝖎𝖓𝖊 🕯
𝖓𝖎𝖓𝖊 🕯@atlanticesque·
Tenant protection laws hit these sorts of places hard. Apartment hotels only work when the landlord can come in with a couple of goons and throw your ass on the street as soon as you’re intolerably late on rent. Hampering the eviction process makes the math just stop working.
Midwest Antiquarian@Eric_Erins

We used to have these. They were called Apartment Hotels. They’d consist of a single room with a bathroom, housecleaning, a cafeteria and lounges. Imagine being able to rent month to month and not need to furnish an apartment. It was ideal. Nuts we got rid of these.

English
43
207
4K
226.6K
Scotty Weeks retweetledi
Nate Silver
Nate Silver@NateSilver538·
I don't consider myself an AI optimist or pessimist particularly. I've gone through different phases relative to consensus. But the "stochastic parrots" people are basically telling the Wright Brothers that their "flying machines" will never work. Just a total embarrassment.
English
57
41
843
111.8K
Scotty Weeks retweetledi
madeofmistake
madeofmistake@madeofmistak3·
at work everyone was uncomfortable with using "master" as the main branch name on git so i changed it to "slave_coordinator"
English
91
455
14.4K
317.9K
Scotty Weeks retweetledi
sigfig
sigfig@sigfig·
people misunderstand the icarus story. the problem was not that he flew too high. it's that the wings were made of beeswax, which offered very little resistance to heating. with modern materials he would have had no problems. we can fly as close to the sun as we want now
English
307
10.5K
122.9K
2.3M

Keşfet

@alexolegimas @avidseries @Musa_alGharbi @elonmusk @BarackObama @taylorswift13 @cristiano @BillGates