[ Built with Jev · Voice control ]
Tool calling that starts before you finish speaking
An experiment replacing the language model's tool-choice step with Jev.
By Pratim Bhosale (@BhosalePratim) ·
Played around with @typesafeai Jev today, mostly to understand what it does for tool calling. Instead of an LLM deciding what to do, I substituted that part with Jev. My learning is that we will be able to make the agent act before the user finishes the sentence. So far,
478likes on X
How it uses Jev
Jev picks the tool instead of an LLM, which the author found lets the agent act before the user finishes the sentence.
It's an example of a wider pattern: see Jev for agent and browser decisions for when it fits, the questions to ask and what it costs.
Summary written by JevHub. Figures are the author's own; we haven't reproduced them. The project belongs to its author.
