Week 7 Curriculum: Building AI Apps

Module 1: Client-server Architecture

By the end of this module, you are able to:

Analyze client-server architecture features, use cases, and examples given a system diagram.
- Identify the appropriate layer (client or server) for a specific unit of code.
Evaluate the benefits of logical separation of concerns in distributed systems.
Explain the necessity of specialized protocols like HTTP, LSP, and MCP.
Trace the HTTP protocol flow and distinguish between Header and Body functions given a raw request.
Differentiate between a Resource and its various Representations.
Deconstruct a URL into its component parts (scheme, host, path, query).
Analyze a RESTful endpoint to identify Nouns (resources) and Verbs (actions).

By the end of this module, you are able to:

Evaluate the key features of FastAPI for building high-performance APIs.
Distinguish between dev and production servers
Implement Pydantic models to validate and serialize data.
Apply Python type hints and docstrings to enhance code quality.
Appreciate the value of well-written documentation, as exemplified by FastAPI docs
Construct a simple RESTful API with multiple resources (items, users) and methods (GET, POST).
Structure a FastAPI application using APIRouter for modular routing.
Utilize advanced Pydantic features including BaseModel, Annotated, Literal, nested structures, and @field_validator.

Distinguish between standard smart IDE features and AI-powered coding assistance.
Classify AI agents based on Interaction and Environment dimensions.
Analyze factors contributing to the operational cost of Coding Agents.
Apply efficiency strategies to enhance development velocity.
Demonstrate proficiency with VS Code navigation shortcuts (Quick file navigation, Go to Definition, Rename symbol).
Demonstrate proficiency with VS Code editing shortcuts (Multi-cursor, Find/Replace, Global Search).
Execute file comparisons using VS Code’s diff tools.
Explain the role of MCP servers in extending Coding Agent capabilities.
Justify the necessity of providing LLMs with up-to-date documentation.

By the end of this module, you are able to:

Construct a minimal MCP Server using the MCP Python SDK or FastMCP.
Refactor an existing FastAPI server into a compliant MCP server using fastapi-mcp.
Integrate an Agent with an MCP server using an MCP Client (Google ADK, LangChain, or Pydantic AI).

By the end of this module, you are able to:

Distinguish between a Container (runtime) and an Image (build-time artifact).
Utilize DockerHub to locate and pull official container images.
Orchestrate multi-container environments using Docker CLI and Docker Compose.
Configure Docker volumes to persist data generated by running containers.
Differentiate between execution stack layers: Hardware, Platform, Runtime, and Application.
Establish a secure database connection using the psycopg driver.
Construct schema definitions and execute database queries using SQLAlchemy.
Manage database schema evolution using dbmate migration scripts.

By the end of this module, you are able to:

Configure a FastAPI application for production by separating configuration settings from code using Environment Variables.
Deploy a web service to a PaaS provider (e.g., Railway) by integrating with a Version Control System.
Provision a managed release of PostgreSQL and configure internal networking to connect it with the web service.
Diagnose deployment errors (e.g., Build failures, Crashing processes) by inspecting platform logs.
Validate the public deployment by executing requests against the production URL.
Implement a Continuous Deployment (CD) workflow where git push triggers automatic rebuilding and redeployment.