📝 SUMMARY This week's roundtable featured Patrick Chouinard demonstrating his Community Brain RAG system that indexes 68 sessions of call transcripts, Juan Torres showcasing an AI photo-booth app for events, and Morgan Cook presenting a Raspberry Pi digital signage kiosk built over a weekend. The group also debated the shifting landscape between Claude Code and Codex, with several members planning migrations due to quality concerns and billing quirks. 💡 KEY INSIGHTS Data preparation beats model selection in RAG pipelines. Patrick found that cleaning and chunking transcripts cost only $0.75 of a $30 total spend, while embedding model choice (Nomic vs Gemini) showed minimal difference when underlying data was solid. Hybrid retrieval is essential for conversational data. Combining BM25 keyword search with vector search significantly outperformed either method alone when querying call transcripts. Small local models can reach 75-80% of frontier quality. With strong retrieval infrastructure, GPT-4o-mini and open-source alternatives approached Opus-level results at a fraction of the cost, though Gemma 4 struggled with complex retrieval logic despite its 30B parameters. Enterprise Claude Code requires configuration hardening. Deploy organizational settings.json files to supersede personal configs, and use seeded CLAUDE.md files to explain constraints rather than just blocking users. Watch for billing mode triggers. Mentioning competitor agent names like OpenClaw or Hermes in Claude Code sessions switches from subscription to API token billing. The Superpower framework now bridges Claude Code, GitHub Copilot, and Codex. This allows workflow portability as members evaluate switching between coding agents. Model default preferences are the new SEO. If a technology isn't in a model's training-weighted defaults, it effectively doesn't exist for AI-assisted development. ❓ KEY Q&A Q: What replaced ShipKit for the group? A: Most have moved to Claude Superpowers, which now works across Claude Code, GitHub Copilot, and Codex.

📞 HOW THE CALLS WORK The calls can run 2+ hours. We want to make sure we're respecting everyone's time. Especially those of you who actually show up. Here's the structure: 👉 Reply to this post with your questions before the call 👉 If you submit a question and you're on the call, you go first 👉 We work through questions in the order they came in 👉 Then we open it up for everyone else I f you can't make the call but want your question answered, drop it in the comments. We'll get to it. But priority goes to people who are there. The goal is simple: if you're taking the time to show up, you shouldn't have to wait behind questions from people who aren't even on the call. 🔗 ZOOM LINK (save this) https://us06web.zoom.us/j/81995207847?pwd=Xe6u6LmIQOmCP5VTnOwWYjDBfZNKGB.1 📅 WHEN This is for Tuesdays March 3rd at 6PM ET Looking forward to seeing you on the calls!

https://github.com/cweagans/n8n-oidc Given that n8n charges $400/month for this, I thought some of you might find it useful to have an alternative.

VIEW RECORDING - 145 mins (No highlights) Key Takeaways - Brandon pre-launched ShipKit, a comprehensive AI development toolkit with templates and resources - Several members reported progress on client projects and personal AI initiatives - Discussions covered cloud deployment, database scaling, and AI personality development - The group shared insights on tools like N8N, LangExtract, and various cloud services Topics ShipKit Launch - Brandon announced the imminent launch of ShipKit, a toolkit for AI application development - Includes templates for chat, RAG, and agent applications in simple and SaaS versions - Features AI-assisted setup and development processes - Aims to simplify the journey from idea to production-ready AI applications Client Projects and Networking - Al Cole secured gigs through networking at business events, focusing on AI automation for marketing agencies - Juan Torres presented at a conference in Istanbul and is working on RAG systems for clients - Jake Maymar is developing projects under NDA, concerned about scaling from 100 to 10,000 users

Here's what I was reading, from the link at the bottom: Why CLI > MCP Agents are really, really good at calling CLIs (actually much better than calling MCPs), so you don’t have to clutter up your context and you can use all the features that Peekaboo has on demand, no installation required. Just add a note in your project’s CLAUDE.md or agent instructions file that “peekaboo is available for screenshots”, or simply mention peekaboo whenever your current context requires visual debugging. As Armin Ronacher perfectly articulates in “Code Is All You Need”, CLIs offer composability, reliability, and verifiability that complex tool interfaces can’t match. CLIs work for both humans and AI agents – we can run, debug, and understand them. Once a CLI command works, it can be executed hundreds of times without requiring additional inference or context. This mechanical predictability makes CLIs the universal, composable interface that bridges human and AI interaction. I’m not saying all MCPs are useless - for example Microsoft’s Playwright MCP for browser automation is great. However, they also built an MCP for GitHub, which is simply a lesser version of the existing gh cli which does the same thing. If this got you thinking, watch Manuel Odendahl’s excellent “MCPs are Boring” talk from AI Engineer. https://steipete.me/posts/2025/peekaboo-2-freeing-the-cli-from-its-mcp-shackles#why-cli--mcp I'm not sure where I fall on this yet, but it sounds perfectly reasonable at a glance.