Gadget-Flow: The Autonomous Smartphone & Laptop Retail Agent

A next-generation Digital Assembly Line for multi-brand gadget retail, powered by Gemini 3.7 and A2A Protocol.

Comment

Inspiration

Managing a multi-brand smartphone and laptop store is a "messy" task. From tracking fluctuating vendor prices, verifying unique IMEI numbers, to securing payments and logistics—it's a multi-step chore that often leads to human error. We wanted to build an agent that doesn't just chat, but takes action to handle the "heavy lifting" of the retail supply chain.

What it does

Gadget-Flow is an autonomous agentic system that manages the entire retail lifecycle. When a new order is received, the system triggers a Digital Assembly Line: Retail Orchestrator (Gemini 3.7): Analyzes the order and decomposes it into task envelopes. Vendor Price Matcher: Autonomously negotiates and matches prices across different distributors. Secure Payment (AP2): Handles autonomous escrow and warranty activation. Smart Inventory: Verifies stock/IMEI and schedules logistics via robotic-conveyor logic.

How we built it

We strictly followed the mandatory requirements for this hackathon: Model: Gemini 3.7 Flash (accessed through Google AI Studio). Framework: Google GenAI SDK for the agentic reasoning and A2A (Agent-to-Agent) communication logic. Infrastructure: Deployed on Google Cloud Run (via Google AI Studio App Hosting) and integrated Google Cloud Firestore logic for real-time transaction logging. Frontend: Built with React, Tailwind CSS, and Framer Motion for a high-tech "Digital Assembly Line" visualizer.

Challenges we ran into

We strictly followed the mandatory requirements for this hackathon: Model: Gemini 3.7 Flash (accessed through Google AI Studio). Framework: Google GenAI SDK for the agentic reasoning and A2A (Agent-to-Agent) communication logic. Infrastructure: Deployed on Google Cloud Run (via Google AI Studio App Hosting) and integrated Google Cloud Firestore logic for real-time transaction logging. Frontend: Built with React, Tailwind CSS, and Framer Motion for a high-tech "Digital Assembly Line" visualizer.

Accomplishments that we're proud of

We successfully moved beyond a "chat loop." Our agent autonomously completes a 6-stage manufacturing and retail process with 100% QA and AP2 settlement verification, all visible through a real-time telemetry dashboard.

What we learned

We learned the power of the Model Context Protocol (MCP) and how Gemini 3.7 Flash can handle complex reasoning for logistics and pricing negotiation far better than traditional hard-coded logic.

What's next for Gadget-Flow: The Autonomous Smartphone & Laptop Retail Agent

Next, we plan to integrate real-world Distributor APIs and physical robot warehouse APIs to turn this digital twin into a fully physical autonomous retail warehouse.

Built With

  • antigravity-sdk
  • gemini-3.7-flash
  • google-cloud-firestore
  • google-cloudrun
  • google-genai-sdk
  • react.js
  • tailwind-css
  • typescript
Share this project:

Updates

Submission history