The network for creativity
Join 1.25M professional creatives like you
Connect with clients, get discovered, and run your business 100% commission-free
Creatives on Contra have earned over $150M and we are just getting started
Back to feedPost
Autonomous RAG-Based AI Customer Support System in n8n
I designed and built an Autonomous RAG (Retrieval-Augmented Generation) AI Customer Support System using n8n, OpenAI, and Supabase to help businesses automate customer support with accurate, knowledge-based responses.
The system is trained on the company's own documentation, enabling it to understand business-specific information and provide fast, context-aware answers without requiring human intervention.
Key features include:
Importing and processing knowledge from business documents.
Splitting and indexing content into a vector database for efficient retrieval.
Using Retrieval-Augmented Generation (RAG) to fetch the most relevant information before generating responses.
Delivering accurate, AI-powered answers based on company documentation rather than generic AI knowledge.
Reducing repetitive support tasks while maintaining consistent, high-quality customer interactions.
This solution is ideal for businesses with extensive documentation, including product guides, knowledge bases, standard operating procedures (SOPs), internal manuals, FAQs, and company policies. By automating knowledge retrieval and response generation, the system improves response times, enhances customer satisfaction, and allows support teams to focus on more complex customer issues.
Built with n8n, OpenAI, Supabase, and Retrieval-Augmented Generation (RAG) to create a scalable, intelligent customer support experience.
The network for creativity
Join 1.25M professional creatives like you
Connect with clients, get discovered, and run your business 100% commission-free
Creatives on Contra have earned over $150M and we are just getting started
Related posts
Spent some time on and off building something just for fun: a live voting site for the argument that never ends — Messi vs Ronaldo.
The football angle is just the hook. Underneath it's closer to a small product than a toy:
- Real payments, real-time — $1 buys a vote, and the global and per-country standings update live as they land
- A 14-tier progression system, Debutant up to GOAT, each level earned and never lost, with three visual card finishes (stands / pitch / history) as you climb
- Every voter gets their own generated card — name, country, camp, vote count — with a shareable page of its own
- A full visual identity built from scratch: stat-heavy and editorial rather than templated
No prizes, no payouts — it's a fan-support board, not a betting site. What I actually wanted to test was whether a dead-simple mechanic (pick a side, back it with money) could carry its own visual language and hold up on repeat visits.
If this one lands, I'd like to keep going with the same format for other sports — there's no shortage of GOAT arguments to settle.
Support and go pick a side: thegoatdebate.io
Might turn this into a proper case study once there's more to show — for now, curious what a design-minded crowd makes of the execution.
Well done! 🚀
Title: AI WhatsApp Booking Automation for Fitness Studios (n8n + Google Gemini AI Agent)
The Problem
The studio was losing members every week without realizing it. Class inquiries came in on WhatsApp at all hours, but staff could only reply during business hours, so by the time someone got a response, they had already booked elsewhere. On top of that, checking real time availability, confirming times, chasing no shows, and logging every booking by hand was quietly consuming hours that the team needed for actually running classes.
The Solution
I built an AI powered WhatsApp automation agent that takes over the entire class booking workflow from first message to post class follow up, cutting the manual work out of the process almost completely. The system combines the WhatsApp Business API, an n8n automation workflow, and a Google Gemini AI agent to run the studio's booking process end to end.
How it works
A member messages the studio's WhatsApp number with a question or booking request. The AI agent reads the message, understands intent, and checks the studio's real class schedule before replying, so every answer reflects actual availability instead of a generic response. It qualifies the lead by asking the right follow up questions (which class, which time, first time or returning member). Once confirmed, the booking is logged automatically to Google Sheets, the studio owner gets an instant confirmation message, and the member receives automated reminders 24 hours and 2 hours before class. If someone doesn't show up, the agent sends a no show follow up on its own, no manual chasing required.
The Result
The studio now handles every inquiry, reminder, and no show follow up in seconds, any time of day, checking real availability instead of guessing, and without adding staff hours or hiring extra front desk support. What used to take a staff member several hours a week now runs on autopilot, which means faster response times, fewer no shows, more classes booked instead of lost to slow replies, and a system that pays for itself many times over in reclaimed time and recovered bookings.
Tools used: WhatsApp Business API, n8n, Google Gemini AI, Google Sheets, Workflow Automation
