I’ve been building an AI workflow for staffing and wanted to share piece by piece how I’m setting it up. Here’s the first piece – Agent 9 (The Intake Agent): Instead of recruiters spending hours digging through inboxes and downloading resumes… 👉 This agent automatically: Collects resumes from emails + job boards Grabs candidate details like name, email, and source Converts resumes into clean text Sends a quick “we got your application” reply Logs everything in one place for the next step It’s basically like having a 24/7 assistant at your front desk. Super simple but it already saves tons of time. ⚡ Next I’ll share Agent 10— where the real magic starts (resume p arsing + structuring).

The creators of agent skills stopped building a separate bot for every single job. In my own AI operating system, I use reusable skills that plug into the same workflow every run. That's why this Anthropic engineer explains what to build instead of AI agents to save you months of wasted time. So what are they actually building? Watch and see.

An Anthropic researcher put his personal estimate of AI extinction risk above 10% within the next decade. I walk through the September 8–13 AI safety debate, Anthropic’s misuse report, Dario Amodei’s plan to slow the race, and Trump’s response. I also share what I think those of us building with and teaching AI can do to help people understand the risks and use these tools responsibly.

Hey everyone, I’m Paul from Canada I’m currently finishing my undergrad in Computer Science, and I joined this community because I want to build my own second brain and create more fluidity in my workflow. I’ve been getting deeper into AI, automation, Claude Code, n8n, and AI agents, but I don’t want to just keep learning new tools for the sake of learning them. My main goal is to figure out how I can use AI to organize the way I think, learn, and work, while building systems that actually make my day-to-day life more efficient. From there, I want to start identifying real problems that I can solve with what I’m learning and turn those solutions into something useful. Excited to learn from everyone here, share what I build along the way, and connect with people who are on a similar path. 🤝

I sat down with Mark Kashef to talk about how to choose the right AI agent and what matters beyond the model. We break down AI harnesses, how we use Claude Code and Codex, and how to keep your skills useful across different tools. We also discuss when to retire old skills and how to keep your AI operating system organized as models improve.