Build a custom AI agent that interacts with users via SMS to perform tasks like matchmaking or scheduling. This advanced workflow uses a low-code stack to connect a database (Airtable), an automation engine (n8n), and a messaging API (Twilio).
with Claire Vo
Workflow automation platform
Cloud-based database and spreadsheet tool
Set up the three core components. Use Airtable as the database for all business data (users, reservations, etc.), n8n as the workflow automation backend (the agent's brain), and Twilio as the user interface (the SMS phone number).
When designing your Airtable base, use flexible free-text fields for user preferences like 'gaming preferences' or 'availability'. This structure is much easier for a large language model to understand and parse compared to rigid, structured data.
Create a new workflow in n8n that triggers when Twilio receives a message. The workflow should first use the sender's phone number to look up the user in your Airtable members table. Then, it passes the user's message to a central AI agent node within n8n.
Give your AI agent in n8n a set of 'tools,' which are simply functions that allow it to interact with your Airtable base. These tools can include actions like create_record, get_records, or update_record.
Configure the AI agent with a system prompt written in plain English. This prompt should explain the agent's purpose, what it should do, and how to use the available tools. The agent will then interpret user requests and figure out which tool to use on its own.
In Twilio, configure your phone number to send an HTTP request to the n8n workflow's webhook URL whenever it receives an SMS. This final step connects your user interface to the agent's brain, making the concierge live.
Hey, can you set up a game for me anytime this weekend? Maybe a deck building game.
Follow Sahil Lavingia's method for delegating small but impactful maintenance tasks, like fixing an internal Slack report, to the AI agent Devin. This workflow shows how to use simple, clear prompts to fix code and update system prompts without distracting your engineering team.
Learn Gumroad CEO Sahil Lavingia's process for rapidly prototyping a UI with v0, then using the AI agent Devin to autonomously code, implement, and submit the feature for review, turning a multi-day task into a one-hour workflow.
Improve your writing in real-time by using ChatGPT as a first reader. Get instant feedback on structure, clarity, and flow as you write, allowing you to polish your work section by section.
We've made ChatPRD affordable so everyone from engineers to founders to Chief Product Officers can benefit from an AI PM.