spencerchubb

For quick and simple, by all means do a TUI. I have done it too, and they're super easy to vibecode :)

Claude Code seems neither quick nor simple

Why do TUI developers insist on doing such weird stuff when they could just make a GUI

use firebase cloud functions free tier

really depends on the company. my company cares a lot about security because it's a huge fortune 50 company with sensitive data and a lot of reputation could be lost with a security scandal

jfrog is the one my company uses

companies that actually care about security have a more secure solution and don't allow devs to use pypi

A human could read the llms.txt if they want to. And a developer could put ads in llms.txt if they wanted to!

You seem to be misunderstanding why a website would make llms.txt

Obviously, they would not make it just for an AI company to scrape

Here's an example. Let's say I run a dev tools company, and I want users to be able to find info about me as easily as possible. Maybe a user's preferred way of searching the web is through a chatbot. If that chatbot also uses llms.txt, it's easy for me to deliver the info, and easy for them to consume. Win-win

Of course adoption is not very widespread, but such is the case for every new standard.

Super exciting that OpenAI pushed the compute out this far

it's even more exciting than that. the fact that you even can use more compute to get more intelligence is a breakthrough. if they spent even more on inference, would they get even better scores on arc agi?

I am eager to learn the pricing as well. It works sooo well but the pricing will make or break whether it's viable for apps

the ability to invert a function sounds crazy to me. I never imagined that was possible

training an ai model is essentially searching for parameters that can make a function really accurate at making predictions. in the case of LLMs, they predict text.

Github already has a way to get the raw text files

Researchers want to publish

Recruiting

if a metric is too simple, choose better metrics! don't throw away metrics altogether

why do you say "just" a wrapper? it is not a bad thing to be a wrapper, it is just a descriptive term

the amount of code does not dictate whether or not it is a wrapper.

I think what you just said actually reinforces the point that ollama is a llama.cpp wrapper. I don't say that to disparage ollama, in fact I love ollama. It is an impressive piece of software. If x uses y under the hood, then we say "x is a y wrapper"

ever since i started working for an insurance company, i realized that understanding the insurance industry is far more difficult than understanding the codebases. if the codebase is acting weird, at least i can step through it with a debugger!

Hunger Games Ballad of Songbirds and Snakes

you don't have to have read the other Hunger Games because it is set about 60 years before the others

not sure what the best is these days because models improve so rapidly. LocalLlama subreddit is probably a good place to ask

How would you determine how much carbon was used throughout the entire supply chain of an imported product? If the product is produced domestically, the government can enforce every business to measure its carbon emissions, but cannot do the same for imported products

1.5 degrees celsius is 10% of the global average temperature. That is, to put it mildly, a shit ton of a difference

It's a strawman to say that money is fake because no serious economist argues that money is "real". Of course it is fake. We use money because it's a useful way to organize society

Of course stopping climate change will save in the long run. The critical question is who should spend the money now to save money later. That is the crux of the free rider problem.

Pricing in externalities is probably the only way we can solve carbon emissions, and would still be very difficult. We would need global participation, otherwise carbon emissions can just be outsourced to other countries. Also, we need to decide the price of emitting carbon. Perhaps survey economists from every country and aggregate the answers somehow.

I think we can all agree that coercion would be bad here. It is currently illegal and should remain illegal.

Hypothetically, if someone wants to sell a kidney to pay the bills, and they are of sound mind, the government shouldn't stop them.

I don't understand this analogy. If someone is dead they cannot use a million dollars

How is it exploitation to legalize more types of consensual transactions?

Theranos was not a "complex business". It was deliberate fraud and deception, and investors that were just gullible. The investors should have demanded to see concrete results

LLMs have powered products used by hundreds of millions, maybe billions. Most experiments will fail and that's okay, arguably even a good thing. Only time will tell which ones succeed