How to Use Lindy AI (Beginner Tutorial for Your First Agent)

About this video

Kevin walks through building a first Lindy AI agent with no coding, using a link that gives $20 in free Lindy credits. He describes the agent in plain language: monitor a Gmail inbox, answer customer questions from a knowledge base if possible, otherwise tell the customer he will follow up and notify him on Slack. In the Flow Editor / Agent Builder he confirms the build, then reviews the Email Received trigger with filters for body text and internal vs external senders, adds a knowledge base from text — the Kevin Cookie Company FAQ covering store hours, whether cookies contain nuts, and contact phone and email — and a condition with two branches. If an answer is found, Lindy auto-generates subject and body via a prompt; the Save as Draft option can be left off for automatic sending. If no answer is found, it emails the customer that someone will follow up and sends a Slack direct message. He tests both paths: a FAQ question gets a reply with store hours, and a question about the world record for fastest cookie dunk follows the no-answer path and triggers Slack. He shows refining agents via the Agent Builder (e.g., adding then deleting a HubSpot ticket step), deploying, monitoring runs in the Tasks tab, exploring templates, and Autopilot / Lindy Computer, where an agent can use a computer to visit websites and log into accounts.

Key takeaways

• Describe the automation in plain language in Lindy’s prompt, then click Build Agent and confirm in the Agent Builder — no coding or if-else logic is required
• The agent is built from a trigger (Email Received), a knowledge base (files, text, or website URLs), a condition, and action steps like Send Email and Send Slack DM
• Test before deploying: a question in the FAQ should take the auto-reply branch, a question outside it should email the customer about follow-up and notify you on Slack
• Use the Tasks tab as an activity feed to see inputs, steps followed, and final output for every agent run, to monitor and debug it
• Refine live agents by typing changes to the Agent Builder instead of rebuilding, and add Autopilot via Lindy Computer Start Computer + Agent Step for computer-use tasks

Episode details: published September 17, 2025 · 12:59 long. Summary written by Self Growth Videos from the episode itself. Business results and any income figures mentioned are Kevin Stratvert’s own claims, shared for education — they are not typical or guaranteed.

← Previous: n8n for Beginners: Build Your First No-Code Workflow (Hostinger VPS Setup) · Next: Genspark AI Tutorial – Meet the AI Super Agent →

All Kevin Stratvert videos · Kevin Stratvert profile · Kevin Stratvert on YouTube