building a computer assistant with Jev local whisper listens to everything I say which then gets classified by Jev to determine what actions to take it uses a small Swift app to provide the full accessibility tree to Jev (i.e. tell its what's on my screen, what can be clicked, etc) still super early, but promising and all real-time which Jev was also local though. then it would be completely private

Everyone figuring out what to build with Jev reminds me of 2020 when we had GPT-3 and we weren’t sure yet what to do with it There was no chat. No turn-taking. No system prompt. No tool calls. All it was, was one large textarea you would prefill with a starting text and some parameters One of the playground examples was a rudimentary chat. To make it work, you’d write some made up chat history, and defined the newline character as the place for the AI to stop generating. Allowing you to then write the next line The below screenshot was from the summer of 2020. I had a lot of fun talking to many celebrities like Eminem 😄 It wasn’t until two years later that OpenAI released a productized version with ChatGPT which of course changed the world Now, when we think of LLMs, we usually think of this turn-taking variety, but back then it wasn’t obvious that this would be the most useful implementation. It was just one demo in a gallery of many Seeing Jev, it makes me wonder what its chat-equivalent is. Is there a killer use case hiding in plain sight that will change the world 2 years from now?