Mudda System Architecture

Microservice Overview of the AI-Powered Civic Engagement Platform

Architecture Diagram

High-Level System Architecture

Microservice Grid

All components collaborate to deliver transparent and trusted civic workflows.

Frontend

Flutter Mobile Frontend

The mobile application used by citizens to report, view, and discuss civic issues.

Frontend

Next.js Dashboard

Admin + Moderation dashboard for issue tracking, AI-assisted resolution, and analytics.

Backend

Spring Backend (AWS Hosted)

Handles authentication, user management, posts, workflows, data pipelines, and integration with all AI microservices.

AI Model

Hate Speech Detector

NLP microservice that flags hate speech, toxicity, abusive content, and policy violations.

AI Model

Post Duplication Detector

AI service that identifies similar or duplicate issues using embedding-based semantic similarity.

AI Model

AQI, Pothole Detection, etc. AI Models

CV + Sensor-based detectors for environment and infrastructure issues.

Agentic Module

Agentic AI Plan Generator

LLM-based workflow engine that creates step-by-step resolution plans for civic issues.

Agentic Module

Workflow APIs

API layer powering automation, escalations, assignment, and state transitions.

RAG Module

RAG

Retrieval-Augmented Generation system that provides context-aware AI responses by querying the civic knowledge base.

Storage

AWS S3

Stores user-uploaded images, videos, and documents.

Storage

PostgreSQL (PSQL)

Main relational database for structured application data.

Storage

Qdrant Vector Database

Stores embeddings for semantic search, deduplication, and similarity matching.

Tech Stack

Core technologies powering the Mudda platform

Spring Boot

Flutter

Next.js

Temporal.io

LangChain + LangSmith

Hugging Face

Qdrant Vector DB

PostgreSQL

AWS Ecosystem