EC1B1 Python Coding SupportFollow-up resources for LSE BSc Economics
← Back to resources

Pedagogical Approach: Reverse Engineering Learning

How we use AI explanations, small tasks, and verification to build durable Python skills.

Updated 2026-01-23

What is reverse engineering learning?

The approach is literally reverse engineering: you start with a piece of code and work backward to understand what it does and why it works. Instead of learning syntax rules first and then writing code, you see working code first and ask AI to explain it to you.

This flips the traditional teaching model:

  • Traditional: Learn theory → Apply rules → Write code

  • Reverse engineering: See code → Ask for explanation → Understand concepts

The reverse engineering learning loop

  1. See a snippet of code (from a notebook, demo file, or example).

  2. Ask AI to explain what the code does, line by line.

  3. Clarify any concepts you still don't understand — ask AI to explain the underlying ideas (e.g., "What is a for loop?" or "Why do we use indexing here?").

  4. Run the code yourself and observe the output.

  5. Modify one thing (a variable name, input value, or loop size) and predict what will change.

  6. Repeat to deepen understanding.

Why this works

Students learn faster when they can see a solution, test it, and explain it. Starting with working code removes the blank-page problem and lets you focus on understanding rather than struggling with syntax from scratch. AI becomes a tutor that answers your questions on demand.

How coding agents fit in

Cursor includes AI features that can speed up coding, but they require active review:

  • Tab completion suggests code as you type.

  • Inline edit lets you rewrite a selected block in natural language.

  • Chat and Agents can explain or propose changes across files.

Agents are powerful, but they are not a replacement for your judgment. Always check outputs and make sure you understand the result.

AI-assisted, human-verified

The guiding rule: you are responsible for correctness. AI can help draft and explain, but you must verify logic, calculations, and assumptions before submitting work.

Practical study habits

  • Do the first exercise without AI, then compare your result with an AI explanation.

  • When stuck, ask for a hint or explanation rather than a full solution.

  • Keep a running note of concepts you do not yet understand and revisit them weekly.

Recommended next steps

  • Complete the Workshop 1 student notebook.

  • Practice in Cursor using the demo files from the session.

  • Use the weekly summaries to track what to review before the next session.