819 karma · joined October 20, 2022
- Another AI firm building products focused on Fortune 500 scale problems. If you're not at a F500, this tool isn't necessarily a good fit for you, so YMMV.
- static analysis tools that produce flowcharts and diagrams like this have existed since antiquity, and I'm not seeing any new real innovation other than "letting the LLM produce it".
They say it's ZDR, so maybe I don't fully understand what problem they're trying to solve, but in general I don't see the value add for a system like this. Also onboarding isn't necessarily just presenting flow charts and diagrams: one of the biggest things you can do to onboard somebody is level-set and provide them with problem context. You COULD go into a 30 minute diatribe about how "this is the X service, which talks to the Y service, and ..." and cover a whiteboard in a sprawling design diagram, or you could just explain to them "this is the problem we're working on", using simple, compact analogies where/when applicable. If the codebase is primarily boilerplate patterns, like CRUD, MVC, or Router/Controller/Service/DB, why talk about them? Focus on the deviant patterns your team uses. Focus on the constraints your team faces, and how you take the unbeaten path to navigate those constraints.
I've also tried the 3 C's, and it still feels like Windsurf has the net best user experience.
My whole fucking point is that neither are the AI scaling laws.
Please stop talking to me.
You're kind of separating yin from yang and pretending that one begot the other. The reason so much money flooded into chip fab was because compute is one of the few technologies (the only technology?) with recursive self improvement properties. Smaller chip fab leads to more compute, which enabled smaller chip fabs though research modeling. Sure: and it's all because humans want to do business faster. But TSMC literally made chips the business and proved out the pure play foundry business model.
> Even if Moore's Law was never a thing
Then arguably in that universe, we would have eventually hit a ceiling, which is precisely the point I'm trying to make against the article: it's a little silly to assume there's an infinite frontier of exponential improvement available just because that was the prior trend.
> Moore's Law has very little to do with the physical size of a single transistor
I mean it has everything to do with the physical size of a single transistor, precisely because of that recursive self improvement phenomenon. In a universe where moore's law doesn't exist, in 2025 we wouldn't be on 3nm production dies, and compute scale would have capped off decades ago. Or perhaps even a lot of other weird physical things would probably be different, like maybe macroscopic quantum phenomena or just an entire universe that is one sentient blob made from the chemical composition of cheeto dust.
We’re at the end of Moore’s Law, it’s pretty reasonable to assume. 3nm M5 chips means there are—what—a few hundred silicon atoms per transistor? We’re an order of magnitude away from .2 nm which is the diameter of a single silicon atom.
My point is, 30 years have passed since dial up. That’s a lot of time to have exponentially increasing returns.
There’s a lot of implicit assumption that “it’s just possible” to have a Moore’s Law for the very concept of intelligence. I think that’s kinda silly.
- lack of technological fluency
- lack of mathematical/scientific fluency
- lack of financial fluency
- lack of political fluency
- lack of critical thinking skills
- willful engagement with streaming psy-op platforms: Facebook, Instagram, Netflix, Tiktok
On an RTX 3080 Ti+Ryzen 9
We’re different people.
--edit--
I'm not referring to his antisemitism. I'm referring to his claim that the Israeli state has their hands in everything. In general I think Kanye West is a fucking moron. But even a mentally unstable schizophrenic clock...
In general, I probably shouldn't have posted this. I'm sorry.
There's one story from Macintosh era where he spent weeks harassing one engineer over the calculator app: "it looks too bloated, it looks afwul, these lines are terrible.." until the engineer got fed up and said "here this is the Macintosh Calculator App: Steve Edition. You get to pick your font, your layout, your color theme." And Steve sat there for hours literally tinkering on a calculator app until he got what he wanted.
He tried to get Paul Rand to change the colors on the NeXT logo--who promptly directed Steve to go have s*x with himself.
There was one point in the Apple Store's inception where they had basically reached done, and Steve decides that he didn't like a certain aspect of it, which was essentially going to require them to redo the entire thing. So he did, and they started from scratch. I think it might have been the Carrara marble floors but I can't remember for certain.
I'm not saying I have immense respect for the guy as a human being, but he was absolutely a notorious tinkerer--a complete menace of one.
Admittedly I don't know that much about LLM optimization/configuration, so apologies if I'm asking dumb questions. Isn't the value of needing to copy/paste that prompt in front of your queries a huge bog on net token efficiency? Like wouldn't you need to do some hundred/thousand query translations just to break even? Maybe I don't understand what you've built.
Cool idea either way!
``` $sum($myArrayExtractor($.context)) ```
where `$myArrayExtractor` is your custom code.
---
Re: "how did it go"
We had a situation where we needed to generate EDI from json objects, which routinely required us to make small tweaks to data, combine data, loop over data, etc. JSONata provided a backend framework for data transformations that reduced the scope and complexity of the project drastically.
I think JSONata is an excellent fit for situations where companies need to do data transforms, for example when it's for the sake of integrations from 3rd-party sources; all the data is there, it just needs to be mapped. Instead of having potentially buggy code as integration, you can have a pseudo-declarative jsonata spec that describes the transform for each integration source, and then just keep a single unified "JSONata runner" as the integration handler.