Saltar al contenido principal

Topic 9: Cloud Deployment Capstone

⏱️ Estimated time: 1 week

Deploy your Journal API from the Phase 2 Capstone to a secure 2-tier cloud architecture. This capstone challenges you to research, design, and implement a production-ready environment with proper networking, security, and cost management.

📚 Learning Path​

Challenge Overview​

Create a secure 2-tier architecture:

  • Tier 1: FastAPI application server (internet-accessible)
  • Tier 2: PostgreSQL database server (private, secure)

Learning Objectives​

  • Design secure cloud networking with public/private subnets
  • Implement security groups following least privilege principles
  • Configure remote database access securely
  • Deploy a production-ready multi-tier application
  • Understand the cost implications of your deployment

Prerequisites​

  • Completed Phase 2 Capstone (Journal API with AI analysis working locally)
  • Cloud provider account with appropriate permissions
  • SSH key pair for server access
  • LLM API key from your chosen provider (OpenAI, Anthropic, Azure OpenAI, AWS Bedrock, or GCP Vertex AI)

Your Mission​

1. Design Your Architecture​

Research and plan your cloud infrastructure:

  • How will you separate public and private resources?
  • What networking components do you need?
  • How will you secure communication between tiers?
  • How much will your deployment cost to run per month?

2. Deploy the API Server​

Set up your application server in the public tier:

  • Choose appropriate compute resources
  • Deploy your Journal API application
  • Configure secure access and networking

3. Deploy the Database Server​

Set up PostgreSQL in the private tier:

  • Ensure it's not directly accessible from the internet
  • Configure it to accept connections from your API server
  • Implement proper security measures

4. Test Your Deployment​

Validate your architecture:

  • All API endpoints work correctly
  • Database is properly secured
  • Data persists between restarts

Success Criteria​

Your deployment succeeds when you can demonstrate:

  • ✅ API accessible from the internet
  • ✅ Database only accessible from API server (not internet)
  • ✅ All CRUD operations work via API endpoints
  • ✅ AI analysis endpoint returns sentiment and summary
  • ✅ LLM API key stored securely (environment variable, not in code)
  • ✅ Data persists between server restarts
  • ✅ Security follows least privilege principles
  • ✅ Budget alerts are configured

Document Your Learning​

Create a technical writeup covering:

  • Your architecture design and reasoning
  • Security decisions and trade-offs
  • Challenges encountered and solutions
  • Key learnings about cloud deployment

Share your progress in the cloud phase Discord channel!

Remember​

This is about understanding cloud patterns, not following a recipe. Research, experiment, and learn from the journey. Every cloud provider has different tools - figure out what works best for your chosen platform.

✅ Topic Checklist​

Before moving on, make sure you have:

  • Designed a secure 2-tier architecture
  • Deployed API server in public tier
  • Deployed PostgreSQL in private tier
  • Database only accessible from API server
  • All CRUD operations working via API
  • AI analysis endpoint returning results
  • LLM API key stored securely
  • Data persisting between restarts
  • Security following least privilege principles
  • Considered costs and configured budget alerts
  • Created technical documentation of your deployment