🎬 Microservices-Based Video Processing Platform

This project is a scalable, microservices-oriented platform for user authentication and secure video processing, built using FastAPI, AWS services (Cognito, S3, SQS, ECS), and FFmpeg.

---

📌 Architecture Overview

---

🧩 Microservices

1. 🔐 Authentication and User Management (FastAPI)

• Uses AWS Cognito for secure user signup, login, OTP confirmation, and token refreshing.
• Stores user metadata (name, email, Cognito sub) in a PostgreSQL database via SQLAlchemy.
• Provides a secure cookie-based session mechanism using access_token and refresh_token.

2. 📦 Consumer Service

• Polls messages from an SQS queue triggered by S3 file uploads.
• Launches ECS Fargate tasks to transcode uploaded videos.
• Reads environment configuration using pydantic-settings.

3. 🎥 Transcoder Service

• Runs inside ECS Fargate and uses FFmpeg to convert video files to adaptive streaming (DASH) formats.
• Downloads video from S3, transcodes to 360p, 720p, and 1080p, and uploads processed videos to another S3 bucket.
• Notifies the backend to update video processing status.

---

🛠 Technologies

• Backend: FastAPI, SQLAlchemy, PostgreSQL
• Cloud Services: AWS Cognito, S3, SQS, ECS Fargate
• Processing: FFmpeg
• Containerization: Docker
• Config Management: pydantic-settings, .env files
• DevOps: Docker Compose (for local DB & FastAPI)

---

📁 Project Structure

. ├── backend/ │ ├── main.py │ ├── routes/ │ ├── services/ │ ├── db/ │ ├── Dockerfile │ └── ... ├── consumer/ │ ├── main.py │ └── Dockerfile ├── transcoder/ │ ├── main.py │ ├── transcode_video.py │ └── Dockerfile ├── architecture_diagram.png └── README.md

---

🚀 Running Locally

1. 🐳 Docker Compose for Backend & DB

docker-compose up --build

# Example for consumer
cd consumer
python3 main.py