Honestly, I've fallen asleep with this trend. Question, have any of you built conversational chatbots for companies using n8n? I've been making them for a while, but just for practice and demos. But at a production level, I haven't done any until now that I have the opportunity to test one. Still, I'm not sure if n8n is ultimately the best option for it. This chatbot needs to answer questions about the company and its programs, and schedule appointments. And I already have a large part of it built. But relying on a single prompt to do all that doesn't seem effective to me. But if I split it into several agents, then it's too many calls, which makes it not very cost-effective. So I'm in that dilemma.

If you’ve been thinking about joining AI Automation Society Plus but the subscription cost has been holding you back, here’s something you should know. There are two ways you can become a member for FREE: 1) Weekly Rewards System - Every week, we reward the top 3 most engaged members in the community: 🥇 $50 for 1st 🥈 $25 for 2nd 🥉 $15 for 3rd - The leaderboard is based on how many likes your posts and comments receive. That means if you finish 1st just twice in a month, your membership is basically paid for—and then some. 2) Affiliate Program - For every person you invite to join AI Automation Society Plus, you earn 40% of their monthly membership for as long as they’re in the group. - Invite just three people and you could cover your own membership indefinitely—and even earn extra each month. So if the only thing holding you back is the ongoing cost, just know, you can earn your way in. You can be paid to participate. And once you're in, you get access to everything, including some brand-new courses dropping in a few weeks. The price is going up soon, so if you’ve been on the fence, now’s your best chance to lock it in. See you inside. Claim your spot here 👉 https://www.skool.com/ai-automation-society-plus/about

Today I was finally able to test the use of MCP Servers—it was giving me some trouble at the proxy level, but luckily I had someone who looked into it for me and helped me fix it. Anyway, here’s the thing: I cloned a Multi-Agent, which basically turned into just one Agent connected to several MCP Servers, and from my point of view, this replaces the Tool Agent that needed to be connected to other agents to be more efficient. Here, I achieved the same result with just one Agent and only the MCP Servers. What do you guys think?

For a deeper discussion on workflow architecture, let's chat here. AI agents are powerful, but here’s the problem—most people are using them when they don’t need to. If your process is deterministic, predictable, and follows a set of structured steps, then an AI workflow is almost always the better option. In this video, I break down the key differences between AI workflows and AI agents, why forcing an agent into a simple process is a mistake, and how switching to a workflow can make your automations cheaper, more consistent, and more scalable. I’ll also show real examples in n8n where I initially built an AI agent, only to realize that a workflow could do the job better.

In this video, I showcase how I built a simple AI-powered Invoice Agent using n8n with no code! I walk through creating a workflow that takes an invoice image sent via Telegram, uses an OCR (Optical Character Recognition) API to extract text, populates the information into a database, and uploads the original image to Google Drive. The agent then provides a quick summary of the invoice, along with the file name and a database link.