visarga

32 ms·

361.
by visarga 9mo ago
Thank you for posting the project, I was actively looking for a solution, even vibe coded a throw away one. One question - how do you pass the credentials for agents inside the cage? I would be interested in a way to use not just claude cod
362.
by visarga 9mo ago
> it's for the same reason that construction workers wear hard hats in environments that will eventually be safe for children. Good response, but more practically, when you are developing a project you allow the agent to do many thi
363.
by visarga 9mo ago
this is gpt, right?
364.
by visarga 9mo ago
> The judge always has to be you. But you can automate much of that work by having good tests. Why vibe-test AI code when you can code-test it? Spend your extra time thinking how to make testing even better.
365.
by visarga 9mo ago
> But how much will it cost to maintain those features in the future? Very little if they have good specs and tests.
366.
by visarga 9mo ago
I see a parallel to how Google search created incentives for SEO and social network feeds created incentive for attention grabbing slop. Platforms optimizing for their own interest at the expense of both upstream and downstream. Is there an
367.
by visarga 9mo ago
The problem is when you use your "copy" as inspiration and actually create and publish something. It is very hard to be certain you are safe, besides literal expression close paraphrasing is also infringing, using world building e
368.
by visarga 9mo ago
Well said, I have been saying the same. Besides helping agents code, it helps us trust the outcome more. You can't trust a code not tested, and you can't read every line of code, it would be like walking a motorcycle. So tests (ba
369.
by visarga 9mo ago
The "pattern matching" perspective is true if you zoom in close enough, just like "protein reactions in water" is true for brains. But if you zoom out you see both humans and LLMs interact with external environments whic
370.
by visarga 9mo ago
But the trend line is less ambiguous, models got better year over year, much much better.
371.
by visarga 9mo ago
> You can't scale your way out of hallucinations You scale your way only out in verifiable domains, like code, math, optimizations, games and simulations. In all the other domains the AI developers still got billions (trillions) of
372.
by visarga 9mo ago
Context is the moat, you can't eat so that I feel satiated, my context my benefits, it is nonfungible
373.
by visarga 9mo ago
> Looking at PRs of agents feels a wrong headed It would be walking the motorcycle.
374.
by visarga 9mo ago
Untested undocumented LLM code is technical debt, but if you do specs and tests it's actually the opposite, you can go beyond technical debt and regenerate your code as you like. You just need testing to be so good it guarantees the be
375.
by visarga 9mo ago
If you implement a project, keep the specs and tests and re-implement it, it should not matter the exact way it was coded as long as it was well tested. So you don't need deterministic LLMs. I think work with LLMs should be centered on
376.
by visarga 9mo ago
TIL about ADR's, a great idea.
377.
by visarga 9mo ago
It's not new, been used like that since the 80's. It scales the logits in a sum of exponentials.
378.
by visarga 9mo ago
> they really, really want to own the entire value chain rather than being relegated to becoming just another "model provider" I remember the story used to be the other way around - "just a wrapper", "wrapper AI
379.
by visarga 9mo ago
A related trick - if you want to teach your agent a specific kind of behavior, and want this behavior to be calibrated and safe, what you can do is: 1. enumerate the actions (policies) your agent takes, collected from prior runs 2. infer t
380.
by visarga 9mo ago
Take responsibility by leaving a good documentation of your code and a beefy set of tests, future agents and humans will have a point to bootstrap from, not just plain code.
381.
by visarga 9mo ago
Learning happens when your ideas break, when code fails, unexpected things happen. And in order to have that in a coding agent you need to provide a sensitive skin, which is made of tests, they provide pain feedback to the agent. Inside a g
382.
by visarga 9mo ago
I don't think "understanding" should be the criteria, you can't commit your eyes in the PR. What you can commit is a test that enforces that understanding programatically. And we can do many many more tests now than befo
383.
by visarga 9mo ago
You don't just code with AI, you provide 2 things 1. a detailed spec, result of your discussions with the agent about work, when it gets it you ask the agent to formalize it into docs 2. an extensive suite of tests to cover every angle
384.
by visarga 9mo ago
> because the crucial change with LLMs is that you don’t use a formal language anymore to specify what you want, and get a deterministic output from that You don't just code, you also test, and your safety is just as good as your t
385.
by visarga 9mo ago
For all these "open questions" you might have it is better to ask the LLM write a benchmark and actually see the numbers. Why rush, spend 10 minutes, you will have a decision backed by some real feedback from code execution. But t
386.
by visarga 9mo ago
Duration of copyright is one way it was perverted, but the other direction was scope. In 1930 judge Hand said in relation to Nichols v. Universal Pictures: > Upon any work...a great number of patterns of increasing generality will fit eq
387.
by visarga 9mo ago
Copying is theft, generating is theft, and it is not even taking anything they had. Future revenue can't be stolen. I think once it becomes infrastructure and widely used knowledge the authors can't claim control anymore. Or shoul
388.
by visarga 9mo ago
This sounds pretty damning, why don't they implement a n-gram based bloom filter to ensure they don't replicate expression too close to the protected IP they trained on? Almost any random 10 word ngram is unique on the internet. A
389.
by visarga 9mo ago
> This isn’t just a health assistant. This is infrastructure for a healthcare marketplace. It's an AI written article IMHO.
390.
by visarga 9mo ago
Talking about embarrassing bugs, Claude chat (both web and iOS apps) lately tend to lose the user message when there is a network error. This happens every day to me lately. It is frustrating to retype a message from memory, first time you

More ›