3 Matching Annotations
  1. Jul 2024
    1. We added an API proposal that enables tools or functions calling. The API comes with two major parts: The ability for extensions to register a "tool". A tool is a piece of functionality that is meant to be used by language models. For example, reading the Git history of a file. The mechanics for language models to support tools, such as: extensions passing tools when making a request, language models requesting a tool invocation, and extensions communicating back the result of a tool invocation.

      Interesting further integration design.

    1. Whoosh provides methods for computing the “key terms” of a set of documents. For these methods, “key terms” basically means terms that are frequent in the given documents, but relatively infrequent in the indexed collection as a whole.

      Very interesting method, and way of looking at the signal. "What makes a document exceptional because something is common within itself and uncommon without".

    1. “If you don’t have your data house in order, AI is going to be less valuable than it would be if it was,” he said. “You can’t just buy six units of AI and then magically change your business.”

      Oh the irony.