🚀 Model Monitoring — Mini

A lightweight ML Observability & Model Performance Dashboard, inspired by TrueFoundry’s real-world monitoring workflows.

This mini-project demonstrates:

Model registration

Version deployment

Live metric streaming

Time-series visualization

API-driven architecture

Frontend dashboards for ML ops teams

⭐ Features

📌 Model Management

Create new ML models

View model list

Switch between models

Version tracking (v1.0, v1.1, etc.)

📈 Live Metrics Dashboard

Real-time charts for:

errorRate

latency

throughput

Color-coded, multi-axis visualization

Time-series updates with each “Push Metric”

⚡ Model Versioning

Deploy new model versions with a button click

Track deployment timestamp

🛰 API-Driven Architecture

Backend provides:

Model storage

Metric computation

Version history

Live metric pushing

Frontend consumes:

/models

/models/:id/metrics

/models/:id/push

/models/:id/version

🏗 Tech Stack

Frontend

React.js (Vite)

Recharts (graphs)

Axios (API client)

Vanilla CSS (light styling)

Backend

Node.js + Express

In-memory store (simulated DB)

Randomized metrics generator

REST APIs

🛠 Local Setup

1. Clone repo

git clone https://github.com/rajeev2004/Model-Monitoring-System.git
cd Model-Monitoring-System

2.▶ Run Backend

cd backend
npm install
nodemon index.js

Backend runs at:

http://localhost:5000

3.▶ Run Frontend

cd frontend
npm install
npm run dev

Frontend runs at:

http://localhost:5173

🌍 Live Demo

🔗 Live URL: https://rajeev2004.github.io/Model-Monitoring-System/