Curious what happens here? Does anthropic charge me for this? Does any of the context generated influence future requests? Or did I just burn 45m and a bunch of usage limits for nothing?
@disconcision When Microsoft created "My Computer" and "My Documents", they demonstrated they did not understand the point at all.
The entire thing is my computer. They are all my documents.
Adam Leventhal and Bryan Cantrill will be talking with @KathiFisler and me on the Oxide and Friends podcast this evening (5pm Pacific/8pm Eastern) about our "experiences enabling students with agentic AI". Join in!
https://t.co/QZTAVHZpqC
Adam Leventhal and Bryan Cantrill will be talking with @KathiFisler and me on the Oxide and Friends podcast this evening (5pm Pacific/8pm Eastern) about our "experiences enabling students with agentic AI". Join in!
https://t.co/QZTAVHZpqC
Part of this is also social, we believe in the community around a given compiler project to use these software engineering techniques, and to accept fixes. This isn't the case with LLMs. (Why did it write a bug? I don't know, Can you prevent it from doing it agin? no)
I think people have begun to use "deterministic" when they mean "explainable" or "testable". When there are bugs in clang/rustc/ghc/whatever we can identify and repair the fault, and use software engineering techniques to help prevent recurrence.
I’m giving a talk at NJPLS (5/22 at Penn)!
I’ll be presenting a DSL for specifying hardware communication protocols & some cool things you can do with it.
Joint work w/ Nikil Shyamsunder, Francis Pham, Adrian Sampson & Kevin Laeufer
https://t.co/aWYe75RgKf