-
-
Ui dashboard
-
invoice Jason created by retail ai agent
-
mongodb atlas order dashboard
-
mongodb atlas products dashboard
-
google console cloud run
🏪 Inspiration
Walking into small retail stores across India, we saw the same problem everywhere — stock tracked in notebooks, bills written manually, and zero analytics. Store owners were losing money daily because they lacked the right tools. We wanted to change that.
💡 What It Does
Retail AI Agent is a fully cloud-native AI assistant that lets any store owner manage their entire store through simple natural language commands.
Just type:
create order for Rahul product Nike Air Max qty 2→ Order created, stock deducted, GST invoice ready!low stock alerts→ Know what to reorder before running outrevenue report→ Real-time earnings dashboardinvoice for Rahul→ Professional GST invoice instantly
🛠️ How We Built It
- Google Agent Builder + Gemini 2.5 Flash → AI brain
- MongoDB Atlas → Real-time database
- MongoDB MCP Server → Bridge between AI and database
- Python Flask → Backend API
- Google Cloud Run → 24/7 deployment
- HTML/CSS/JS → Frontend dashboard
The MongoDB MCP Server is the core of our integration — it enables real-time database operations through natural language commands.
🚧 Challenges We Faced
- Making natural language parsing accurate for retail commands
- Auto stock deduction and restoration on order create/delete
- GST invoice generation with correct tax calculations
- Deploying everything on Cloud Run with zero downtime
📚 What We Learned
- Building production-grade agents with Google Cloud Agent Builder
- Integrating MongoDB MCP Server with a Flask backend
- Designing customer-facing vs admin-facing agent architecture
- Cloud Run deployment with environment variables and auto-scaling
🏆 What's Next
- Razorpay/Stripe payment gateway integration
- PDF invoice download
- WhatsApp order notifications
- Multi-store support
- Customer login and order history
Log in or sign up for Devpost to join the conversation.