Build an AI-Powered Tech Radar Advisor with SQL DB, RAG, and Routing Agents

AI-Powered Tech Radar Advisor This workflow builds a sophisticated AI assistant that acts as a "Tech Radar Advisor." It allows users to ask natural language questions about the company's technology stack, strategic direction, and adoption status across different business units. The system intelligently determines the best way to answer the query, either by searching a structured database or by performing a contextual search on documentation, and then delivers a polished, safe, and relevant response. Technologies Used - Data Sources: Google Sheets, Google Drive - Databases: MySQL (for structured, queryable data), Pinecone (as a vector store for semantic search) - AI Platforms & Models: - Google Gemini: Used for creating the numerical representations (embeddings) of the tech radar documentation. - Groq: Provides high-speed inference for the Large Language Models (LLMs) that power the agents (Deepseek, Llama-3). - LangChain: The underlying framework used within n8n to structure the AI agents, tools, and chains. - Core Automation: n8n for workflow orchestration and integration. - User Interface: A Webhook endpoint allows any front-end application to interact with the AI advisor. What It Automates & The Value Provided This workflow automates the entire process of creating and maintaining a hybrid AI search system, providing significant value: - Automated Data Synchronization: It automatically keeps both a traditional SQL database and a modern vector database perfectly in sync with a single source of truth (a Google Sheet). This eliminates error-prone manual data management and ensures the AI always has the most current information. - Intelligent Q&A System: It transforms a static spreadsheet into a dynamic, conversational advisor. Employees no longer need to manually parse complex tables; they can simply ask questions like, "Which new tools are strategic but not yet adopted by Company 2?" and get an immediate, accurate answer.