@pro.glitch: An entire AI front office for a plumbing company, piece by piece. Every piece is something you can learn to assemble. ine at night, office closed. A customer fills in one form. n8n picks it up, Claude reads it like a good receptionist (drain clearing, postcode in range, details complete → qualified), booking gets a time, sales gets the appointment, the service team gets the job, the CRM keeps the history, the job tracker shows the owner. A reschedule moves the visit and cancels the old reminder. A quote request goes to a human in the team chat with full context. The owner sees one view: bookings, follow-ups, work awaiting approval. What that system actually is: 1. Trigger + fixed path = automation 2. Claude reading and deciding = AI inside the automation 3. Service and postcode rules = the gate 4. Booking, CRM, chat = connections with permissions 5. Customer records = the business's second brain 6. The quote going to a person = human in the loop 7. The owner's view = the report The thing to take from this was never this exact workflow. It's understanding automation, where AI fits in a path and what it's doing at every step, how to route a thing, when to add a human, when to let it run. Once you know the blocks, a plumbing office, a dentist, a cleaning company, a law office are just different arrangements, and any workflow becomes something you can draw and then build. #aiautomation#aiagent

Professor Glitch

Open In TikTok:

Region: US

Friday 25 September 2026 20:00:00 GMT