ECO Analyzer is a retrieval-based analytics platform with an AI Copilot that lets non-technical users query large internal datasets in natural language.
The Copilot (GPT-4) grounds every answer in the platform's own data — 18,977+ doctors, 247 teams, and 688 cities — instead of hallucinating from model memory. The same RAG pattern is reusable: it can wrap any internal data source and expose it to teams who would otherwise need an analyst or SQL.
How it is built:
• AI Copilot — natural-language interface backed by GPT-4 with retrieval grounding, so answers come from the data, not from the model's imagination.
• Microservices — 8 independently deployable services (FastAPI + Node.js), each owning one slice of the domain.
• Data layer — PostgreSQL with PostGIS for geospatial queries across 688 cities, Redis for caching.
• Retrieval — a reusable RAG pattern for internal knowledge access, designed to be pointed at new data sources without rebuilding the pipeline.
Skills: RAG, AI Copilot, GPT-4, Microservices, FastAPI, PostgreSQL, PostGIS, Redis, Python, Node.js