🤖 NodeMesh — AI-Powered Intent-Based Assistant

NodeMesh is a Stateful Micro-Orchestrator designed to minimize inference latency while maximizing contextual depth through a multi-layered middleware architecture. It transitions from traditional linear request-handling to a concurrent pre-processing pipeline, ensuring that the AI isn't just reacting, but proactively analyzing every interaction.

Features

• Sarcasm, Intent-Based Query Handling
  Automatically detects user intent or sarcasm across:

  • general/sarcasm → Groq llama-3.3-70b-versatile
  • weather, rain, alert → WeatherMap API
  • news → NewsAPI

• Smart Prompt Engineering
  Each API is guided by tailored prompts to ensure clarity, relevance, and structured output. Gemini is used for summarization, reasoning, and fallback logic. This handles the sarcasm of the user and answers it rather than assuming it literally. Apart from this, we have incorporated Gita Shlokas to answer distressed and mentally tired queries(used few-shot learning)

• Frontend

  • Built with React + Vite
  • Responsive UI with Dark/Light Mode toggle
  • Clean routing and modular component design

• Backend

  • Runs via npm run server
  • Handles API orchestration and intent parsing
  • Secured with .env for API key management

---

🧠 Powered by Gemini Flash 2.0

Groq llama-3.3-70b-versatile is used to:

• Handle general queries with natural language understanding
• Synthesize Substantive Data: Transforms raw, "ugly" JSON payloads from WeatherAPI and NewsAPI into coherent, contextually rich narratives.
• Structure Responses: Automatically organizes info using Markdown for style, Bullet Points for lists (like news or activities), and JSON to double-check its own logic.

Prompt engineering is reflected in how each query is structured and routed, ensuring high-quality, context-aware responses.

---

Tech Stack

Layer	Tools Used
Frontend	React, Vite, Tailwind CSS
Backend	Node.js, Express
APIs	Groq llama-3.3-70b-versatile, WeatherAPI, NewsAPI
Dev Tools	Postman, ESLint, dotenv

---

Testing the APIs

Use Postman to test backend endpoints by importing the API sample code/ pasting it in the raw format area and request using "POST" with your API Key to verify the working.

🛠 Setup Instructions

1. Clone the repo

   git clone https://github.com/ANKVIT26/NodeMesh_Assistant.git
   cd NodeMesh_Assistant

2. Install dependencies

   • The project has both frontend and backend folders.
   • Run:

     npm install

     inside each folder (frontend and backend).

3. Create .env file

   • In the root or backend folder, add a .env file with your API keys:

     GEMINI_API_KEY=your_gemini_key
     WEATHER_API_KEY=your_weather_key
     NEWS_API_KEY=your_newsapi_key

4. Start the backend server

   npm run server

5. Start the frontend

   npm run dev

   • This launches the React + Vite frontend with hot reload.

⚡ Tech Stack Recap

• Frontend: React, Vite, Tailwind CSS
• Backend: Node.js, Express
• APIs: Groq llama-3.3-70b-versatile, WeatherAPI, NewsAPI
• Dev Tools: Postman, ESLint, dotenv

#DEMO VIDEO https://drive.google.com/file/d/1Z-AfodYTV25SY84mCoV4Bwn10EC85z0G/view?usp=sharing

React + Vite(Official guide template)

This template provides a minimal setup to get React working in Vite with HMR and some ESLint rules.

Currently, two official plugins are available:

• @vitejs/plugin-react uses Babel for Fast Refresh
• @vitejs/plugin-react-swc uses SWC for Fast Refresh

React Compiler

The React Compiler is not enabled on this template. To add it, see this documentation.

Expanding the ESLint configuration

If you are developing a production application, we recommend using TypeScript with type-aware lint rules enabled. Check out the TS template for information on how to integrate TypeScript and typescript-eslint in your project.