Ja’dan Johnson

6.4K posts

Ja’dan Johnson

Ja’dan Johnson

@jdanjohnson

🏗️ Creative Technologist. Prev. @devpost | Building technical communities. 🚀Catalyst @MiamiHackWeek @osmedsupplies ✨Vibes Curator ⛓️ Connector 🔧 Builder

Earth Katılım Mart 2011
999 Takip Edilen5K Takipçiler
Eli Mernit
Eli Mernit@mernit·
For the past few months, I’ve been running an AI swat team. We learn about a company, audit their workflows, and build them an AI app to automate their busywork. This sounds straightforward, but building these apps is annoying because you keep rebuilding the same things over and over: a filesystem, sandboxes per user, auth, streaming, and connections to messaging apps like Telegram and Slack So today we’re releasing all of it together in a batteries-included Python framework called Capsule It includes a sandboxed computer, multi-channel messaging, background jobs, multi-user auth, and payments – everything you need to ship a production-ready AI product super fast The faster you ship, the more you’ll learn from real customers, and the sooner you get PMF
English
14
4
68
7.2K
Ben
Ben@bensenescu·
I'm open sourcing Downy: a team of self-improving, OpenClaw-esque, agents that you can work with from any device. Talking to your agents through iMessage, Discord or the terminal is a hack. Downy is self-hosted, but it doesn't run on your machine. Instead, you deploy it into your Cloudflare account and leverage their Agent Infrastructure. Each agent only has access to the workspace and tools you give it in the Cloud, instead of root access to your entire life. It works on the Free Plan if you use your OpenAI Subscription or OpenRouter. Downy is self-editing like OpenClaw or Hermes. You can ask it to do research and create skills, or to add MCP servers. These are then immediately available to the agent without needing to restart or edit any files. Downy is built for many focused agents instead of trying to teach one agent or system how to do everything. Each agent gets its own: - identity - memory - workspace - skills - tools So you can have a product agent, a marketing agent, a research agent, a content agent, or whatever else you want. Downy's goal is to make working with a team of self-improving agents feel pleasant instead of like a crazy science experiment. Hope you check out the GitHub, give your feedback and maybe even a star!
English
12
7
125
11.5K
Ja’dan Johnson
Ja’dan Johnson@jdanjohnson·
Teaching the team to Devin-maxxxx!
Ja’dan Johnson tweet media
English
0
0
7
1K
Nathan Baschez
Nathan Baschez@nbaschez·
Do you spend a lot of time reviewing markdown docs written by AI? Wish it were a better experience? Say hi if you wanna try a new (free, open source) thing
English
341
1
353
55.1K
Gui Grupenmacher
Gui Grupenmacher@gui_grup·
I wouldn't believe it if you told me something like this would be possible just a few months ago. If you're also building an AI-powered GTM Ops team like we are, reach out, would love to exchange notes (and tell you about how we build with @DevinAI)! 🤩
English
1
0
3
78
Ja’dan Johnson retweetledi
Cognition
Cognition@cognition·
The terminal hasn’t changed much since the 1970s. What you do with it has. Introducing Devin for Terminal: everything we learned building Devin, now as a local agent, available right in your shell. And when your work outgrows your laptop, hand it off to the cloud.
English
58
119
1.1K
359.3K
Ja’dan Johnson
Ja’dan Johnson@jdanjohnson·
@grinich How have you enabled this though? How do you get nontechnical folks to correctly architect usable products? Genuinely curious
English
0
0
0
237
Ja’dan Johnson
Ja’dan Johnson@jdanjohnson·
Okay I can’t be the only one wishing I could build more in community. Anyone engineers working on AI & Agents in Miami would be down to cowork together? Looking to start a small group to nerdmaxxx
English
7
2
38
2.4K