This is a submission for the Algolia MCP Server Challenge
What I Built
Supply-Chain Query Dashboard — a smart, AI-enhanced logistics dashboard that helps businesses and individuals track delayed shipments across warehouses using natural language queries.
This tool empowers users to:
Ask intuitive questions like "Where is my parcel?" or "Show all delayed orders from Warehouse B over 2 days".
Filter by warehouse, delivery status, or delay duration.
Visualize shipment origins on an interactive map.
Get intelligent delay summaries and AI-categorized product types using integrated backend enrichment.
Built with Streamlit, Algolia MCP Server, Flask, and AI agents (OpenAI) to deliver a seamless, fast, and intelligent user experience.
Demo
Github - https://github.com/shivamkumar123321/Supply-Chain-Dashboard
How I Utilized the Algolia MCP Server
This project leverages the Algolia MCP Server in multiple powerful ways:
Natural Language Query Translation
Queries like "delayed over 2 days from A" are processed through the MCP server and converted into Algolia-compatible filters (facetFilters, numericFilters, etc.).AI Agent Proxy Layer
A custom mcp_proxy.py Flask server receives requests, forwards them to MCP, and then pipes the structured response directly into Algolia’s search.Dynamic Search Experience
The frontend interacts live with the MCP-backed Algolia API, ensuring fast, semantic, and precise data retrieval.
Key Takeaways
Building with MCP: Learned how powerful Algolia’s MCP server is when combined with open-ended user input.
AI Enrichment: Integrated Claude + OpenAI for real-time classification and summary generation.
Frontend–Backend Coordination: Created a proxy pattern (mcp_proxy.py) to ensure clean API usage between UI and MCP.
Challenge: Balancing real-time filters with flexible NLP required careful query translation and testing
💥 This project could easily scale into a full logistics SaaS product for e-commerce vendors, warehouse operators, or shipping companies!
Top comments (0)