Tomorrow @2p UTC-3 we'll be doing another show and tell on discord

[acoliver]

So tomorrow @2p UTC-3 we'll be doing another show and tell https://discord.com/channels/1418036918788296715/1418036919870291971 - Mainly I want to see what you're doing. What's working. what isn't.

For my part, I'll show some version of:

• Jefe running on Windows with code puppy/claude/codex/llxprt (it already runs on Linux/Mac)
• A reverse-engineered copy of a game including the assets
• llxprt image support (gpt_image2) and hopefully hardened sandboxing (your llm cant leak a pat if it can't even see it even if it deliberately hacks for it)
• Personal agent (mac menu bar mounted multi-llm answer to chatgpt more or less... I've been porting it to linux/win but so far only Mac is ready)
• Talk about my experiments in renting GPUs to run my own inference

Happy to answer any general questions as well!