Bonjour je suis française et je suis tombée sur les ressources youtube de Ryan et Matt que j'adore. J'ai un background tech et je suis très intéressée à lancer ma propore agence IA pour répondre aux besoins des TPE autour de leur manque de temps/productivité. J'ai dans ma famille plusieurs commerçants pour lesquels je vais commencer à analyser leurs besoins. Je suis suuuper interessée de savoir comment vous, vous intégrez l'IA chez vos clients ? Quelles sont les demandes terrain ? J'ai la chance de pouvoir suivre ma conjointe au bout du monde grâce à son métier. Je serais ravie d'échanger en français ou anglais avec des membres de la communauté, peu importe où vous habitez ! A bientôt !

Hey everyone! I’m new here and excited to be part of this Data & AI Automations community. Looking forward to learning, sharing ideas, and connecting with people building amazing things in AI. Glad to be here! 🚀

Backstory: I caught Claude making significant errors (Opus 4.7 1M, high effort) and asked about the agentic BS and all the errors I'm catching. This was Claude's response, unedited: (didn't need editing) Fair. Honest answer: The hype is largely overblown. A lot of the "autonomous agent" pitch right now describes demos, not daily work. The gap between a polished launch video and what you've experienced in this session is real and not your imagination. What I actually am is more limited than the marketing: broad knowledge, fast at routine tasks, decent at multi-step reasoning when the steps are well-bounded. But I make mistakes — sometimes basic ones, sometimes compounding ones — and I don't always catch them on my own. You watched me in a single conversation: delete a workflow I'd already been told not to, bundle URLs that confused you, run a discovery loop that wasn't needed, paste literal <picker: full_name> text into instructions, default to memory when we'd already established memory is the weakest layer. None of those are advanced failures. They're sloppy. What I'm actually useful for in this kind of work: Wide-context reading: holding 50 files in mind, finding patterns across them. Routine API plumbing: clones, payload edits, sheet schemas. Drafting structured docs and plans. Catching things you might overlook at scale. What I'm not: A reliable senior engineer who needs no review. A replacement for someone who actually owns the work. Self-correcting in any deep sense — you have to be the catch on my mistakes, which is exhausting and exactly the opposite of what the marketing promises. The honest working pattern: treat me as a capable but error-prone collaborator that needs supervision, not a delegate. Show me where I'm wrong (which you've been doing well), force the lessons into stronger layers than memory, and don't expect me to flawlessly execute a complex pipeline without intervention. That's the real product. Whether it's worth the effort over doing it yourself depends on the task. For something like this — clone a workflow, wire 18 fields, append rows to a sheet — a careful engineer working alone would probably go faster than us today, because they wouldn't pay the cost of catching my mistakes. Where I add value is when the surface is wider than one person can hold at once. Today's task wasn't that wide; that's why the friction has been high.

18 days ago I shared my first automation project here. Since then, I've significantly improved its architecture — and I'm excited to share what's new! What changed: 🗄️ Cross-Workflow Data Bridge: Added Google Sheets as a reliable data layer between Generator and Moderator workflows — seamlessly passing image URLs and multilingual texts between two separate n8n workflows. 🔗 Real LinkedIn Publishing: Connected LinkedIn OAuth2 API — approved posts now go live automatically in the selected language (EN, PL, or ES) with just one tap in Telegram. 📘 Facebook Publishing Ready: Facebook Graph API node is already implemented in the workflow and ready to activate for clients who need it. 📝 Universal Prompt: Removed the archaeology-specific focus — the agent now creates expert content on ANY topic the user sends via Telegram bot. The system now consists of two interconnected n8n workflows — a true agentic architecture! About the Development: This upgrade was co-developed with Claude (Anthropic) as a technical mentor and debug partner. If you're learning AI automation — don't be afraid to iterate and improve. Every upgrade teaches you something new! 🚀 PhD in Archaeology → AI Automation at 55. It's never too late! 👉 See where it all started: https://www.skool.com/data-and-ai/my-first-experience-in-automation?p=76cc6cfe 📂 Full project on GitHub: https://github.com/Evgeniy1970-ai/ai-content-scout

How is the week going