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
My Lovable × Contra challenge entry is taking shape: Wagging Wagon — a mobile dog-grooming business run by 7 specialist AI agents. Route-aware booking, AI coat/size assessment from a photo, by-size pricing, and a live owner console over all of it. The hard part wasn't the interface. It was the handoffs — seven agents passing a booking from quote to dispatch to groomer without dropping context. That's the whole game in multi-agent systems. Full walkthrough before submission. 🐾
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
Lupe's Mobile Grooming (Cicero, IL, fictional): turning "can I book with you?" into "you're booked" for a one-van, one-groomer dog groomer who can't answer the phone with wet hands.
The problem: the van works one area a day, a visit takes 1 or 2 hours depending on the dog, 1 in 8 visits fails at a locked gate, and the waitlist lives on paper.
What it does (Spanish and English):
- Front door: the dog's size and coat set the visit length; the ZIP code only offers days when the van is in that area; gate code and parking are collected before arrival; instant confirmation with a calendar file and a private link to confirm, reschedule or cancel.
- Follow-through: at 6 pm every customer for the next working day gets a check ("someone home? can Lupe get in?"). At 7 pm unconfirmed visits are released, and each freed slot goes to the longest-waiting person in that area whose dog fits. Cancellations refill the same way. The owner does nothing; the demo clock on /owner lets you run her evening yourself.
How it was built: one Lovable prompt made the first version. The scheduling rules were then written in code with an AI coding assistant through Lovable's GitHub sync (23 tests). Lovable's security scan flagged open database tables, so every read and write now goes through validated server functions, and the public key is denied.
Fictional business, simulated messages; the booking flow is real.
When a visit is released at 7 pm, does the longest-waiting customer in that area get it booked outright, or an offer they have to accept before the van leaves in the morning?
Wag & Wheels is a fictional one-van mobile dog grooming business in Newcastle upon Tyne, run by Lucy.
The problem: every “can you fit Milo in?” used to mean multiple messages, checking the diary, working out travel time, confirming the service, chasing a deposit, and then rearranging the day if someone cancelled.
I rebuilt that flow so customers can answer a few questions, get appointment options that fit both the dog’s needs and Lucy’s existing route, pay a £20 deposit, and receive an instant confirmation without any back-and-forth.
On Lucy’s side, the booking drops straight into her route, notes are saved, reminders are scheduled, and cancelled slots can be matched to suitable customers on the waitlist.
From “Can I book?” to “You’re booked”, with almost no manual admin for the owner.
#lovablechallenge
Hi Everyone,
I am currently open to new opportunities and actively looking for remote full-time, freelance, or contract Full-Stack / AI Engineering roles.
I'm Krish, a Full-Stack Developer specializing in AI agent systems, and I build products end-to-end. Here are my recent projects and my portfolio:
Portfolio: https://krishchhabra.com
Orchestra — Multi-Agent Coding WorkspaceAI agents that plan, code, test, and open PRs autonomously in disposable sandboxes — you watch it happen live.
Live: https://workspace.krishchhabra.com
Demo (vision): https://live-pr-pilot.lovable.app
Agentra — AI Agent Retrieval ConsoleRAG + MCP-based research agent built on LangChain and LangGraph. Features a multi-agent report generator, live tool execution, persistent memory, and source citations on every answer.
Live: https://agentra.krishchhabra.com
My Stack:React, Next.js, Node, TypeScript, Redis, React Native, SwiftUI, Postgres, MongoDB, Docker, AWS — plus the full LLM ecosystem (OpenAI, Claude, LangChain, LangGraph, MCP, RAG).
If you have any openings at your company, can refer me, or know anyone hiring in the agentic AI space, I'd really appreciate an intro! Happy to share my resume on request.:
