@claudeai Love this. I'm a full-time teacher and dad of 3 who's never written code, and I've been building my own homemade take with Claude Code. Cortana brings what needs me to my phone, keeps working while I teach, and checks with me before she acts. Joining the waitlist to see yours!
@aksenHQ A relay we built. Phone web app โ my Mac over Tailscale โ a small server pings Cortana's Claude Code session, and she answers back to my phone. ๐ซก
Claude Code is a single coordinator in my system. Just get it on the job of connecting all your models. I sent a prompt from my custom Android app straight to my Claude Code terminal on the MacBook. She sent a message to a work thread in the ChatGPT desktop app on my Studio!
The way I use agents (Oct 21)
Currently My AI usage is split between codex and Claude Code, but I use them both differently than I did a month ago.
The way i use codex is mostly through my dot. Every morning I call my dot and put my AirPods in and I walk have a conversation for about 45 minutes.
I try and handle as much work stuff as I can during that conversation. I respond to emails, write instructions for my editors, and do any planning for the day. Small tasks my dot does it, longer tasks I ask it to spin up a codex session.
I REALLY wish I didnโt have to dial and wait for my dot to answer @OpenAIDevs
When Iโm at my desk doing coding work, document creation, or video editing I use Claude code. But, Iโm not using normal Claude code Iโm using Claude Code projects.
The same way dot is a codex coordinator, Claude projects is a Claude code coordinator. Both allow you to spin up cloud sessions and local sessions.
If I need to use computer use I will ask dot to fire up a codex session on my MacBook (that is always on) to do a task. Codex computer use is in leave of its own.
For coding work or editing work that requires my MacBook Iโll ask my claude projects.
Right now I have many Claude Code Projects forโฆ each project. Sometimes I wish it was a single coordinator that could manage and organize all my workโฆ like the way I use dot.
Both Claude and Codex have a grokbot skill for anything Twitter related. So if any task involve x, they will delegate to Grokbot which is very useful as a creator.
But the fact that I canโt see which model is being used when keeps me from using Grokbot. I know I can have Grokbot delegate to cursor which can specify a model but cursor is very expensive when using frontier models compared to codex and Claude, so I donโt end up doing that.
Claude Code is a single coordinator in my system. Just get it on the job of connecting all your models. I sent a prompt from my custom Android app straight to my Claude Code terminal on the MacBook. She sent a message to a work thread in the ChatGPT desktop app on my Studio!
[SOUND ON] Under 30 minutes, zero typing. A 4-minute voice memo to Cortana, my Claude Code agent, turned into a full voiceover studio for my whole CC LABS crew. You can start building stuff like this too, just by asking. ๐๏ธ Polished cut + Raw Demo! ๐ซก
@rileybrown I have my agents end the night by generating a dream journal! A dreamy take on one of the most interesting/important things we did that day. But I LOVE this idea of having them explore new ideas overnight. I've been doing that manually. Time to automate it! ๐ซก Good stuff, Riley!
Every morning Claude Sends me a video now.
I tell it to explore interesting video styles on the internet before making it. It has access to my voice via elevenlabs. I told it to make it 15-30 seconds.
And i give it feedback every morning and it's instructed to pick a new style and improve the next day. here's what it sent me this AM.
Ok, but do they show you how to make your graph Halo themed? And do they show you how to put a Claude code brain in your graph that looks like 343 Guilty Spark and actually flies around the graph connecting nodes and telling the story of your query? ๐ ๐ซก
Ok, but do they show you how to make your graph Halo themed? And do they show you how to put a Claude code brain in your graph that looks like 343 Guilty Spark and actually flies around the graph connecting nodes and telling the story of your query? ๐ ๐ซก
Google just dropped a 1-hour breakdown of Graph Engineering โ showing how to go from one AI agent to a system that runs 24/7.
00:00 โ Understanding graph fundamentals
09:16 โ Build your first AI agent
21:15 โ The concept of Graph Engineering
41:03 โ Putting Graph Engineering into practice
52:21 โ Building self-improving graphs
Itโs completely free and one of the most valuable resources for anyone looking to understand where AI agents are heading.
Prompts โ Agents โ Loops โ Graphs
Most people will watch the first 9 minutes, build a basic agent, and call it a day.
But the final 20 minutes reveal the bigger picture.
Same model.
Same tokens.
A fundamentally different level of capability.
Watch it today.
Love this! If I had kids, Iโd love to teach them this way.
Iโve been exploring learning through play too! A Claude skill for making hand-drawn, interactive everyday objects. Hereโs my little washing machine :)
https://t.co/ImifIlE39K
Hereโs a prototype of the Kids version: โHow does a pump fill a tyre?โ
Iโm experimenting with different iterations to find the simplest way to help kids understand how things work, while still keeping enough detail to make the experience engaging, interactive, and educational
@claudeai notice me senpai :S , im building a full fledged agentic game engine + editor, that will make claude agent a bigboss in computer graphics :D ; it runs both native and on the web , which means literally you could open a web tab inside claude code and just open the engine and voila DONE
Anthropic's most recent IG post about everything they've released this month, summed up by my custom Claude Code agent, Cortana in 30seconds! The post is like a half hour old
@claudeai