Shane

@reustle

3 Following

46 Followers

Shane pfp

That's one way I was thinking, you give it a week or something and it monitors, but feels fragile and doesn't adapt to changes easily. I'm going to spend more time on this https://github.com/orgs/modelcontextprotocol/discussions/84

1 reply

0 recast

2 reactions

Shane pfp

There are a few routes to go down w/ that. What I've been brainstorming for a few weeks, heavily related to MCP, is how to essentially let these agents go hunt down the API docs for the service & your API key when logged in, then it can do most (?) things w/out fumbling w/ the browser.

2 replies

0 recast

3 reactions

Shane pfp

Thanks! 3 in parallel, 4o realtime audio model (basic listening) + whisper model (transcripts) + o3-mini (double-checking regularly)

0 reply

0 recast

3 reactions

Shane pfp

Important clarification: everything is happening on the client side except for the the audio being proxied to OpenAI, then receiving tool calls back.

0 reply

0 recast

2 reactions

Shane pfp

I built a prototype showcasing how I think AI assistants will interact with existing software during an OpenAI Realtime Hackathon this weekend. Won 2nd place and $15k in credits. (skip to 1:00) https://hirebellhop.com

10 replies

11 recasts

50 reactions

Shane pfp

Ahh, nobody was supposed to see this. The submission process required a social media post 😅 I'll post the demo video on here / X shortly.

0 reply

0 recast

1 reaction

Shane pfp

I built a new idea at the @openai realtime API hackathon with @aitinkerers

1 reply

5 recasts

14 reactions