Just added a new free module to the classroom. Most people use Claude daily without checking what happens to their conversations. This guide covers what you need to know and what to change. Check it out here: https://www.skool.com/build-with-ai-9352/classroomc994c7e9md=fb08f47577ec4ae4a92a19b70106510b

When I interned at Atlassian last summer, AI was blowing up in the coding space. A lot of my peers were afraid to touch it. I took the opposite approach. I used Cursor and our in-house AI assistant Rovo for as much of the coding as I could. That freed me up to focus on the stuff that actually mattered - structure, architecture, system design. The stuff AI still gets wrong. I spent a lot of my time talking to Claude. Asking it for ideas, getting direct feedback, treating it like a second opinion. Claude is often wrong, but it usually has the right direction. So I'd guide it, it'd guide me, and together we'd land on a working architecture and ship it. Over the three months I was there, AI evolved fast. I was using it way more by the time I left than when I started. And it was obvious - Atlassian wanted us using these tools. They want their engineers leveraging AI to move faster. By going all in from day one, I finished my project in half the time it took the other interns. That's what stood out. That's exactly why I built this community - to show you how I use AI day to day and how it's made me significantly more efficient as an engineer. If you have any questions about using AI in a real engineering environment, drop them in the comments. Happy to share what I've learned.

Drop a comment below with: 1. What you do, student, developer, founder, something else entirely 2. What you're building right now, or what you want to build 3. Where you're at with AI coding agents, never used one, dabbling, daily driver Make sure to tick off the check box when complete! I'll go first: I'm Nic. Software engineering student at the University of Auckland, AI automations engineer at Allexive, joining Atlassian in 2027. I build AI systems and then show exactly how they work. Right now I'm shipping automations with Claude Code every day and documenting the whole process on Instagram and YouTube (@nonacnic).

Just dropped all my guides into Skool. Free. Hacks, Stacks, Builds, Prompts: full versions, not the previews on the site. Claude Code setup, agent-browser, Shannon, gstack, token saving, debugging prompts, Beads, skills, VoxCPM2. Classroom tab. Check it out.

Vercel Labs dropped this on GitHub and it's quietly replacing Playwright MCP for anyone running agents that browse. Native Rust CLI, daemon architecture, accessibility-tree snapshots instead of full DOM dumps. Same browser, less bloat. What it does differently: - Native Rust daemon, no Node.js or Playwright dependency - Semantic element refs (@e1, @e2) from accessibility snapshots, no brittle CSS selectors - CDP direct, no translation layer - Multi-session, Chrome profile reuse, auth persistence built in Install in 30 seconds: 1. 'npm install -g agent-browser' 2. 'agent-browser install' 3. 'agent-browser start' Then 'navigate', 'snapshot', 'click @e1'. Done. Repo: github.com/vercel-labs/agent-browser. Apache 2.0 licensed. Things to know: - It's pre-1.0 (v0.26), so pin your version - Chrome-only via CDP, though Lightpanda is supported as a lighter alternative - The 93% number is for typical browse-and-act flows, your mileage will vary If you're running agents that browse, this changes the math on your token budget. If you hit anything weird, drop it below. I've probably seen it.