AI CEO

WatonGroup
N/A
Product & Technology Lead
Built AI Q&A, announcement interpretation, and other capabilities Integrated multiple models and knowledge bases to improve response speed

Project Overview

This project aims to provide users with an intelligent personal assistant service. It kicked off in August and has been continuously iterated and refined.

Technical Architecture

  • Backend framework: FastAPI
  • Programming language: Python
  • Web frontend: JavaScript
  • Database: MySQL, Redis
  • AI models: GPT, Claude, Mistral, Kimi, Qwen ∫∫

Key Features

1. Multi-Model Integration

The project integrates multiple large language models, including:

  • Domestic models: Qwen, Zhipu, DeepSeek
  • International models: Claude, the GPT family
  • Multimodal models: MiniCPM-Llama3-V and more

2. Data Encryption and Privacy Protection

The project features a well-designed data encryption and privacy protection mechanism:

  • Phase one: fixed-key encryption
  • Later versions: more advanced encryption schemes

Core Modules

  1. ai-service: the core AI service module
  2. ai-secretary: the secretary features module
  3. ai-image-generator: the image generation module
  4. message-service: the messaging service module