ML Model Registry & Deployment Dashboard

A web application for data scientists and ML engineers to register, version, track, and deploy machine learning models through an intuitive dashboard.

Problem Description

The Challenge

Data science teams face significant challenges when managing machine learning models in production:

1. Model Versioning Chaos - Models are often stored in scattered locations (local files, cloud storage, shared drives) with inconsistent naming conventions, making it difficult to track which version is deployed where.

2. Lack of Metadata Tracking - Critical information about models (training data, hyperparameters, performance metrics, dependencies) is frequently lost or poorly documented.

3. Deployment Friction - Moving a model from development to production typically requires manual intervention, custom scripts, and coordination between data scientists and DevOps teams.

4. No Central Source of Truth - Teams lack visibility into what models exist, their status, and their deployment history.

The Solution

The ML Model Registry & Deployment Dashboard provides:

• Centralized Model Repository - A single place to register and store all ML models with their metadata
• Version Control - Track model versions with automatic versioning and comparison capabilities
• Rich Metadata - Store and query model information including:
  • Model name, description, and tags
  • Framework (scikit-learn, TensorFlow, PyTorch, etc.)
  • Performance metrics (accuracy, F1, RMSE, etc.)
  • Training dataset information
  • Author and creation timestamps
• Deployment Management - Track deployment status and history across environments (staging, production)
• Simple UI - Intuitive dashboard for non-technical stakeholders to view model inventory

Target Users

• Data Scientists - Register models, track experiments, compare versions
• ML Engineers - Deploy models, monitor deployment status
• Team Leads - Get visibility into the ML model inventory and deployment pipeline

Features

Core Features (MVP)

• Register new ML models with metadata
• Upload model artifacts (pickle files, ONNX, etc.)
• List and search models
• View model details and version history
• Update deployment status (development → staging → production)
• Dashboard with model statistics

Future Enhancements

• Model performance comparison charts
• A/B testing support
• Automated deployment to Kubernetes
• Model lineage tracking
• API key authentication
• Webhook notifications

System Architecture

┌─────────────────────────────────────────────────────────────────┐
│                         Frontend (React)                        │
│                                                                 │
│  ┌──────────┐  ┌──────────┐  ┌──────────┐  ┌──────────────────┐│
│  │Dashboard │  │Model List│  │Model     │  │Register Model   ││
│  │  View    │  │  View    │  │Details   │  │     Form        ││
│  └──────────┘  └──────────┘  └──────────┘  └──────────────────┘│
│                              │                                  │
│                    ┌─────────▼─────────┐                       │
│                    │   API Service     │                       │
│                    │ (Centralized)     │                       │
│                    └─────────┬─────────┘                       │
└──────────────────────────────┼──────────────────────────────────┘
                               │ HTTP/REST
                               ▼
┌─────────────────────────────────────────────────────────────────┐
│                      Backend (FastAPI)                          │
│                                                                 │
│  ┌──────────────────────────────────────────────────────────┐  │
│  │                    OpenAPI Contract                       │  │
│  │  POST /models     GET /models      GET /models/{id}      │  │
│  │  PUT /models/{id} DELETE /models   GET /models/stats     │  │
│  └──────────────────────────────────────────────────────────┘  │
│                              │                                  │
│  ┌──────────────────────────┼──────────────────────────────┐   │
│  │              Service Layer (Business Logic)              │   │
│  └──────────────────────────┼──────────────────────────────┘   │
│                              │                                  │
│  ┌──────────────────────────▼──────────────────────────────┐   │
│  │              Repository Layer (Data Access)              │   │
│  └──────────────────────────┬──────────────────────────────┘   │
└──────────────────────────────┼──────────────────────────────────┘
                               │
                               ▼
┌─────────────────────────────────────────────────────────────────┐
│                        Database                                 │
│            SQLite (dev) / PostgreSQL (prod)                     │
└─────────────────────────────────────────────────────────────────┘

Tech Stack

Component	Technology	Purpose
Frontend	React + TypeScript	Modern, type-safe UI
Styling	Tailwind CSS	Utility-first styling
Backend	FastAPI (Python)	High-performance async API
Database	SQLite / PostgreSQL	Flexible data persistence
ORM	SQLAlchemy	Database abstraction
API Docs	OpenAPI 3.0	Contract-first development
Containerization	Docker + Docker Compose	Consistent environments
CI/CD	GitHub Actions	Automated testing & deployment
Cloud	Railway	Production hosting

Project Structure

ml-model-registry/
├── README.md
├── AGENTS.md                    # AI development documentation
├── docker-compose.yml
├── .github/
│   └── workflows/
│       └── ci-cd.yml
├── docs/
│   └── openapi.yaml            # API contract
├── frontend/
│   ├── Dockerfile
│   ├── package.json
│   ├── src/
│   │   ├── components/
│   │   ├── services/
│   │   │   └── api.ts          # Centralized API calls
│   │   ├── pages/
│   │   └── App.tsx
│   └── tests/
├── backend/
│   ├── Dockerfile
│   ├── requirements.txt
│   ├── app/
│   │   ├── main.py
│   │   ├── models/
│   │   ├── routes/
│   │   ├── services/
│   │   └── database/
│   └── tests/
│       ├── unit/
│       └── integration/
└── requirements.txt

Getting Started

Prerequisites

• Docker and Docker Compose
• Node.js 18+ (for local frontend development)
• Python 3.11+ (for local backend development)

Quick Start with Docker

# Clone the repository
git clone https://github.com/HighviewOne/ml-model-registry.git
cd ml-model-registry

# Start all services
docker-compose up --build

# Access the application
# Frontend: http://localhost:3000
# Backend API: http://localhost:8000
# API Docs: http://localhost:8000/docs

Local Development

Backend

cd backend
python -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate
pip install -r requirements.txt
uvicorn app.main:app --reload

Frontend

cd frontend
npm install
npm run dev

API Documentation

The API follows a contract-first approach. See the OpenAPI specification for the complete API contract.

Key Endpoints

Method	Endpoint	Description
GET	/api/v1/models	List all models
POST	/api/v1/models	Register a new model
GET	/api/v1/models/{id}	Get model details
PUT	/api/v1/models/{id}	Update model
DELETE	/api/v1/models/{id}	Delete model
GET	/api/v1/models/{id}/versions	List model versions
POST	/api/v1/models/{id}/versions	Create new version
GET	/api/v1/stats	Get dashboard statistics

Testing

# Run all tests
docker-compose run --rm backend pytest
docker-compose run --rm frontend npm test

# Run with coverage
docker-compose run --rm backend pytest --cov=app

Deployment

The API is deployed at: https://ml-model-registry-production.up.railway.app

Link	Description
API Docs	Interactive Swagger documentation
Health Check	API health status
Stats	Dashboard statistics
Models	List all models

AI-Assisted Development

This project was built with AI assistance. See AGENTS.md for detailed documentation on:

• AI tools and workflows used
• Prompting strategies
• MCP (Model Context Protocol) integration

Contributing

1. Fork the repository
2. Create a feature branch
3. Make your changes
4. Run tests
5. Submit a pull request

License

MIT License - see LICENSE for details.

---

AI Dev Tools Zoomcamp 2025 - Project 1