e1g

10 ms·

61.
by e1g 1y ago
Is there a form of this argument that could not be made leading up to the Industrial Revolution? Approximately 60-80% of pre-industrial ways of making a living disappeared or required complete reskilling. Does anyone seriously think we shou
62.
by e1g 1y ago
Almost certainly a couple of whales as they made the investigation happen.
63.
by e1g 1y ago
As an immigrant, I understand the deal (for a working visa): I must bring valuable and scarce skills that this economy lacks (as demonstrated by the premium it is willing to pay). "Total Compensation" is an imperfect-but-good-enou
64.
by e1g 1y ago
Here's a bulletproof working visa policy: automatically approve everyone who will be in income tax (state and federal) more than the median gross income (~$60k). Sort of minimum-income-tax personal guarantee, which all highly-skilled m
65.
by e1g 1y ago
OpenAI explicitly said it’s not GPT-5 but another experimental research model https://x.com/alexwei_/status/1946477756738629827?s=46
66.
by e1g 1y ago
Yes, that, and AI Explained [1] on YouTube - no BS hype, strong yet approachable analysis, and an in-depth private community if you want a deeper dive. https://www.youtube.com/@aiexplained-official
67.
by e1g 1y ago
It's impossible as an IC on a team, or working where a concept of "tickets" exists. It's unavoidable as a solo founder, whether you're building enterprise systems or expanding your vision. Some details - 1. Before w
68.
by e1g 1y ago
Context rot. My use case is iterating over a large codebase which quickly grows context. All LLMs degrade with larger context sizes, well below their published limits, but pro models degrade the least. R1 gets confused relatively quickly, d
69.
by e1g 1y ago
I’m a founder/CTO of an enterprise SaaS, and I code everything from data modeling, to algos, backend integrations, frontend architecture, UI widgets, etc. All in TypeScript, which is perfectly suited to LLMs because we can fit the type
70.
by e1g 1y ago
Yes, a lot of usage, I’d guess top 10% among my peers. I do 6-10hrs of constant iterating across mid-size codebases of 750k tokens. CC is set to use Opus by default, which further drives up costs. Estimating productivity gains is a flame
71.
by e1g 1y ago
Yes, my CC usage is regularly $50-$100 per day , so their Max plan is absolutely great value that I don’t expect to last.
72.
by e1g 1y ago
Parallel test time compute is exactly what SOTA models do, including Claude 4 Opus extended, o3 Pro, Grok 4 Heavy, and Gemini 2.5 Pro.
73.
by e1g 1y ago
Agree, current "thinking" models are effectively "re-run this question N times, and determine the best answer", and this LLM-evaluating-LLM loop demonstrably leads to higher quality answers against objective metrics (in
74.
by e1g 1y ago
She’s a VC-backed founder who went to Yale, and her very first job was at Goldman. What she’s describing in the article is not “low status” because she hadn’t experienced that. But the feeling she describes reveals what she thinks “low st
75.
by e1g 1y ago
Okay, I just tried the Claude Code extension in VSCode - the experience is 100% identical to how it works within IntelliJ, even using the same icons and shortcuts to turn it on/off.
76.
by e1g 1y ago
I use IntelliJ (WebStorm) as my daily driver, and Claude Code is an integral part of my workflow. I didn’t check VSCode specifically, but I did read the link to that extension - everything they describe is precisely how the Claude Code work
77.
by e1g 1y ago
While your observation is generally true, and I share your overall concern about my IDE of choice, in this specific example it doesn’t apply as the Claude Code plugin for IntelliJ offers exactly the same integration as their plugin for VSCo
78.
by e1g 1y ago
SublteCrypto is async, and the author specifically said they want their API to be sync.
79.
by e1g 1y ago
FYI, you can add browser support by using noble-hashes[1] for SHA256/HMAC - it's a well-done library, and gives you performance that is indistinguishable from native crypto on any scale relevant to S3 operations. We use it for our
80.
by e1g 1y ago
+1 to this concern from our Very Large customers in the EU. Today, AWS, GCP, etc. are omnipresent, so there are plenty of counterexamples; however, the growing concern is, "How do we become less reliant on AWS in the next decade?"
81.
by e1g 1y ago
+1. We use Zod only for internal validation and would love a standalone release for v4 (or another package like zod4 or @zod/zod). I'm sure many greenfield projects would prefer that, too.
82.
by e1g 1y ago
If the business got more benefit, they would be fighting to keep this setup - and none are.
83.
by e1g 1y ago
The prime mover behind this project is Daniel Kokotajlo, an ex-OpenAI researcher who documented his last predictions in 2021 [1], and much of that essay turned out to be nearly prophetic. Scott Alexander is a psychiatrist , but more releva
84.
by e1g 1y ago
This is comparable to another research/estimate with similar findings at https://ai-2027.com/ I find the proposed timelines aggressive (~AGI in ~3 years), but the people behind this thinking are exceptionally thoughtfu
85.
by e1g 1y ago
Especially when unions are involved. We have a flexible schema for many parameters (e.g., X accepts an object or an array of them, and one object has some properties that could be an enum or a detailed shape), and both Zod and Valibot produ
86.
by e1g 2y ago
“Malloy” is one one in the same spirit, but probably more comprehensive https://github.com/malloydata/malloy
87.
by e1g 2y ago
From the last 20 years of bootstrapping to various levels of success, I now have an absolute rule: when I feel like I definitely cannot take a break _now_ that the time I must take a break, non-negotiable. Even a few days is enough. I have
88.
by e1g 2y ago
I don't know of any actively maintained, but here's a drop-in replacement for you that does the same thing with the latest LibreOffice https://gist.github.com/eugene1g/74172a761e604d875a59774189a...
89.
by e1g 2y ago
LibreOffice also can be used headlesslsy to convert other files to PDF locally. We use it via a container to first generate well-styled docx files (letters, invoices etc) and then convert those to PDFs.
90.
by e1g 2y ago
No, I don't know enough math to test the logic, only the check questions against their expected answers in https://anonymous.4open.science/r/putnam-axiom-B57C/data/Put...

More ›