Skip to main content

Infra Basics

Fullstack Dev Environment

Goal​

Bridge the gap between AI development and production. Learn to wrap models in APIs, create web interfaces, and deploy containerized applications.

Estimated Time Impact

12-17 hours total

Web Stack(5h)
FastAPI(5h)
Docker/DevOps(4.5h)

1. The Web Stack​

Core Concepts​

video
beginner

HTML, CSS, JS Crash Course

3-4 hours
video
beginner

DOM Manipulation Tutorial

1-2 hours

Understand the technologies that power web applications to build interfaces for your AI models.

Key Topics:

  • HTML/CSS: Structure and styling of web pages.
  • JavaScript: Interactivity and API calls.
  • DOM: Dynamically updating content without page reloads.

Action Items:

  1. Build a simple HTML form for user input.
  2. Use JavaScript to handle form submission.
  3. Display mock results dynamically using DOM manipulation.

2. Serving AI with FastAPI​

Core Concepts​

video
intermediate

FastAPI Full Course

4-6 hours

Transform Python AI models into production-ready web services.

Key Topics:

  • FastAPI: Endpoints, Path Operations, Pydantic Models.
  • Integration: Loading models efficiently (startup event).
  • CORS: Allowing frontend communication.

Action Items:

  1. Create a FastAPI app with a prediction endpoint.
  2. Load your trained model once at startup.
  3. Connect your frontend from Module 1 to the API.

3. DevOps & Deployment​

Core Concepts​

tutorial
beginner

Docker 101

3-4 hours
video
beginner

SSH Practical Tutorial

1-2 hours
video
intermediate

SSH Concepts: How It Works Behind the Scenes

1-2 hours

Package your application for consistency and deploy it to the cloud.

Key Topics:

  • Docker: Containers, Dockerfiles, Docker Compose.
  • SSH: Secure remote server management.
  • Deployment: Cloud servers (DigitalOcean/AWS) vs. PaaS (Railway/Render).

Note: For deeper networking concepts essential for secure deployments, see the Systems & Networking module.

Action Items:

  1. Write a Dockerfile for your FastAPI app.
  2. Build and run the container locally.
  3. Deploy to a cloud provider (e.g., DigitalOcean Droplet or Railway).

Dockerfile Example:

FROM python:3.10-slim
WORKDIR /app
COPY requirements.txt .
RUN pip install -r requirements.txt
COPY . .
CMD ["uvicorn", "main:app", "--host", "0.0.0.0", "--port", "8000"]

Completion Checklist​

  • Build an HTML page with interactive JavaScript.
  • Create a FastAPI app with Pydantic validation.
  • Serve an AI model via HTTP API.
  • Build and run a Docker container.
  • Deploy to a cloud server (or PaaS).

Capstone Exercise: Build and deploy a complete AI app (Frontend + Backend + Docker).

Common Pitfalls​

  • CORS Errors: Add CORSMiddleware to FastAPI.
  • Slow Predictions: Load models at startup, not per request.
  • Large Images: Use .dockerignore and slim base images.
  • Connection Issues: Check security groups and port mappings.