Live data from Hacker News

Show HN: Agent Draw: An agent draws while you talk, built on TLDraw

techstackups.com

1–10 of 18 posts

Show HN: Agent Draw: An agent draws while you talk, built on TLDraw

#1
We recently built 2draw, a Drawful-style game where players draw on a shared canvas and race to guess each other's drawings, on tldraw, an infinite-canvas SDK for React.

We started wondering what it would take to put an agent in that loop, as an opponent or a rival guesser, and dug into how an agent could read and draw on a tldraw canvas. That research turned into:

Agent draw, a tool that lets an agent draw to the canvas for you while you present.

You can try it right now, or grab the source:

- Live demo: https://tldraw-agent-draw-demo.james-664.workers.dev

- Source: https://github.com/ritza-co/tldraw-agent-draw-demo

Here you can see the agent assisting me in my demo presentation of a third grade chemistry class:

- https://www.youtube.com/watch?v=iIH2hJAxxm8

Show HN: Agent Draw: An agent draws while you talk, built on TLDraw
techstackups.com

Re: Show HN: Agent Draw: An agent draws while you talk, built on TLDraw

#3
A very interesting way of interacting with an AI agent. I tried out the demo for creating mind maps, and it was an interesting experience to see it grow and be able to explore different branches. An interesting concept for exploring AI-supported visual thinking.

Re: Show HN: Agent Draw: An agent draws while you talk, built on TLDraw

#8
This is sick! I built a similar tool at https://shimmerdiagrams.com

I really enjoyed reading your blog post, it was pretty informative about the techniques. I had built mine as a prototype and specifically interested in running all models local and in browser.

Tldraw seems to be a better medium for this that mermaid.

Re: Show HN: Agent Draw: An agent draws while you talk, built on TLDraw

#10
post #6

Very cool. Takes some getting used to but its powerful

Yes it could get a lot better with some design improvements, maybe live streaming the transcription, and maybe after some trial and error testing the best + fastest model for the job
Post reply on HN