Skip to main content

AI Coding Course

Master AI-assisted software engineering for experienced developers

For experienced engineers with 3+ years professional experience

Start Learning
prompt.md
Debug the failing test in UserService.test.ts:

INVESTIGATE:
1. Read the test file, identify which assertion fails
2. Trace code path through UserService
3. Examine error messages and stack traces

ANALYZE:
4. Compare expected vs actual behavior
5. Identify the root cause

Provide conclusions with evidence:
- File paths and line numbers (`src/auth/jwt.ts:45-67`)
- Actual values from code (`timeout: 3000`)
- Specific function names (`validateToken()`)

Interactive Course Content

Diagrams and visualizations from the lessons

Lesson 3

Iterative AI Workflows

Research, Plan, Execute, Validate - systematic AI-assisted development

Lesson 8

Context Isolation

Separate contexts prevent bias between code, tests, and triage

Code QualityTime / Iterations →StartPOSITIVE FEEDBACKNEGATIVE FEEDBACKIter 110 filesIter 212 filesIter 314 filesIter 18g / 2bIter 26g / 4bIter 34g / 6b
Positive: Good patterns compound into better code. Each iteration strengthens quality.
Negative: Bad patterns compound into worse code. Each iteration degrades quality.
Lesson 11

Code Pattern Compounding

Good patterns amplify quality, bad patterns compound technical debt

Course Structure

Module 1

Understanding the Tools

  • LLM internals: context, attention, token limits
  • What breaks: hallucinations, code drift, refactoring
  • Context management and RAG integration
Module 2

Methodology

  • Prompt structure: constraints, examples, chain-of-thought
  • Grounding: embedding context that persists
  • Iteration patterns: plan, execute, verify
Module 3

Practical Techniques

  • CI integration and automated review patterns
  • Test generation and coverage strategies
  • Debugging sessions: when AI makes it worse

Prerequisites

✓ You should have

  • 3+ years professional software engineering experience
  • Access to a CLI coding agent (Claude Code, Aider, Cursor, or similar)
  • Solid understanding of data structures and design patterns
  • Experience with system design and architectural decisions

✗ Not required

  • Prior AI or machine learning experience
  • Deep knowledge of transformer architectures
  • Specific programming language (examples use TypeScript, Python, etc.)

Get Started

Begin with Understanding the Tools to learn how LLMs and coding agents work

Start Learning
View on GitHub•Join Discussions