NEO-Z Chatbot

A full-stack AI chatbot and multimodal workspace with text, image, file, voice, and hardware-interaction capabilities.

May 2026• Team: Personal Project• Role: Full Stack Developer

About this Project

NEO-Z is a cinematic AI workspace and chatbot application. It uses a Node.js backend with MongoDB for auth/chat storage, OpenRouter for text generation, Gemini for image analysis, and an optional Python microservice for voice/audio tasks.

Tech Stack

React (Vite)

Tailwind CSS

React Markdown and remark-gfm

Web Speech API and Web Serial API

Node.js and Express

MongoDB

OpenRouter (google/gemma-4-31b-it by default)

FastAPI

Docker

SpeechRecognition and Pydub

Tools Used

VS Code

Git

Key Features

Core Features

▸Cinematic Hacker UI: Custom dark-themed interface built with React and Tailwind CSS.
▸Real-Time Streaming: Text responses stream from OpenRouter with Server-Sent Events (SSE).
▸Hardware Protocol: Native Web Serial API support for connected hardware via custom <CMD> tags.
▸Data & Vision Processing: Upload images for visual analysis or attach CSV files for bulk data interpretation.

Technical Implementation

▸AI Chat: Streaming text generation through OpenRouter using hosted language models.
▸Conversation Management: Authentication and persistent chat history backed by MongoDB.
▸Multimodal Input: Image analysis and CSV/document-style data interpretation.
▸Voice & Hardware: Optional Python audio service and Web Serial API integration.

Installation

Clone the Repository

git clone https://github.com/QaziZaid16/neoz-ai-chatbot.git

Setup Core Backend

cd backend, npm install, # Create a .env file with MONGO_URI (or MONGODB_URI), OPENROUTER_API_KEY, GEMINI_API_KEY, and JWT_SECRET #, npm start

Setup Frontend

cd frontend, npm install, npm run dev

Optional Python Microservice

cd python-service, pip install -r requirements.txt, uvicorn main:app --reload

Document Q&A System