NTU SC4052 Cloud Computing Project

Mr.Assist
Personal Assistant Platform

Mr.Assist is a cloud-native Personal Assistant-as-a-Service platform that decomposes assistant capabilities into independent microservices. Connect GitHub, Mastodon, and AI-powered task management through well-defined RESTful APIs.

6
Microservices
AES-256
Encryption
17+
Tests Passing
8
Cloud Principles

Six Independent Microservices

Each service is independently deployable, scalable, and communicates via typed RESTful API contracts.

GitHub Integration

Manage repositories, track issues, and monitor pull requests through a unified interface.

Mastodon Integration

Access your timeline, schedule posts, and manage notifications across the Fediverse.

AI Task Suggestions

Get intelligent task recommendations powered by LLM analysis of your workflow patterns.

AES-256-GCM Encryption

API tokens are encrypted at rest with unique initialization vectors per credential.

Activity Logging

Comprehensive audit trails tracking every action across all connected services.

Analytics Dashboard

Track usage metrics, API call volumes, and performance indicators in real time.

Cloud Computing Principles

Mr.Assist demonstrates eight core cloud computing principles through its architecture and implementation.

Microservices Independence
Loose Coupling via APIs
High Cohesion
Independent Scalability
API-First Design
Encryption at Rest
Observability & Logging
Stateless Services

Technology Stack

Frontend

  • React 19 with TypeScript
  • Tailwind CSS 4 for styling
  • tRPC for type-safe API calls
  • Recharts for data visualization
  • shadcn/ui component library

Backend

  • Express.js with tRPC
  • MySQL / TiDB database
  • Drizzle ORM for type-safe queries
  • AES-256-GCM encryption
  • LLM integration for AI features