I made a full walkthrough on how to build a RAG based support agent using n8n, Firecrawl, Supabase and OpenAI. Sharing it here because RAG sounds complex, but the setup is actually pretty simple once you see the steps. Here’s the quick breakdown of the workflow: 1. Scrape the website with Firecrawl It grabs every page and formats everything into clean markdown. 2. Upload the files to Drive Just to store all the scraped content in one place. 3. Pull the files into n8n Download each file so n8n can process them. 4. Convert markdown into text Supabase needs plain text before embedding. 5. Create a Supabase vector store Run one SQL snippet and your table is ready. 6. Embed and upload n8n chunks the text, embeds it and sends it to Supabase. 7. Connect a Telegram bot This becomes the front end for talking to the agent. 8. Build the agent The agent pulls relevant info from Supabase and answers based on what’s actually in your data. The video walks through every step if you want to build one yourself. If you want more step by step tutorials like this, please subscribe to my channel. I will be posting more in the upcoming weeks :)

A picture is worth thousands words. So here it is. Turned on the AI cold calling system for one of my real estate client yesterday. Just 110 calls, 2 active sellers interested to talk to the agent. Each deal is minimum worth 500K USD if it gets closed but that is beyond the scope of AI system so can't comment. All this while Real Estate agent was busy dealing with his clients. Cherry on the cake, the second lead was in hospital while talking and AI wished her speedy recovery with the open ended prompts I built, *** Partial Transcript from Retell AI*** (Can't share the audio for privacy reasons and just changed the names only) User: About a month's time. I'm in hospital at the moment. 1:03 Agent: I’m sorry to hear that you’re in hospital, Sophi. Wishing you a smooth and speedy recovery. We can definitely arrange for John to call you in about a month’s time. Is there a particular day or time that works best for you then? 1:07 User: Not really. Anytime should do. 1:22 Agent: Thank you for your time, Sophi. We'll have John reach out to you in about a month's time to discuss your property appraisal. Wishing you all the best with your recovery. ****************** It is not the smartest or most complex automation out there. Does not even have any calendar booking but adding hell lot of value. So don't chase those 50+ module N8N and make.com screenshots. Build simple yet powerful automation with 20% energy and spend rest of the 80% energy to find clients, try in real world and eventually sell. "1m USD worth leads on the second day from cold calling"

Here's a classic project. It makes it easy to organize emails... You know, because it does it for you. Sorts 'em into four tiered categories: 1) Critical 2) Growth 3) Operational Archive 4) Noise Main thing I learned is to not get too hung up on one solution for getting a data flow to work. For instance, with the 'Review and tag email' flow, AI reviewed the emails, however I wasn't able to immediatly sort them via Gmail. So I split it out into four separate paths... and Filter nodes made that transition work cleanly. (Is there a cleaner way to do this? Would love to hear suggestions.) Thanks for taking the time to check this out. Have an awesome weekend!

Welcome fellow Accelerant, a free community where experts, creators, and builders come together to master AI building, automation, and scalable systems. 🔥 Inside Accelerants, you'll find: ✅ Exclusive Blueprints & Resources – Learn how to build powerful AI automation systems. ✅ AI-Driven Growth Strategies – Insights from top industry operators in every business vertical. ✅ Live Workshops & Expert Insights – Stay ahead with real-world AI applications. We cover everything—from no-code tools for beginners to advanced AI workflows, code generation, and AI-driven platform development. Whether you're a startup, SMB, or enterprise, this is where you'll level up. 📌 Meet the Experts Inside Accelerants (So far... More to come!)Each expert shares resources & insights in the classroom: - Craig Cecilio – Capital raising & investment strategies for AI Founders. - @Michele Torti – SME’s Make. com workflow automations. - Phil Sergenti – Email outbound & inbound automation strategies. - Paolo Trivellato – LinkedIn acquisition & content strategies. - Spencer Donaldson – AI-driven marketing & growth. - Syed Hussain – Enterprise AI automations. - Diego Mayor – AI-driven business scaling systems with Airtable + Make. - Matthew Cohn – AI Voice Agent systems for agencies & SMBs. - Steven DeCesare – AI & systems for branding & creative industries. - Agastya Gudipati – AI sales videos & content workflows. - Marquise Peters – AI YouTube content systems. - Andrew Athanassopoulos – Automation systems for home service businesses. 🚀 More collaborators will be added soon—this is just the beginning! Each expert will introduce themselves in their own post, sharing what they’re bringing to the community. 👇 Now it’s your turn! Drop a comment below and introduce yourself: - Who are you? - What are you building or working on? - What do you want to learn or contribute? - Let’s build something incredible together.

Building out a Miro board for planning workflows in n8n. Feel free to use it / copy it. https://miro.com/app/board/uXjVIVLMOgE=/?share_link_id=823749743273