All in One View
Content from Prompting effectively
Last updated on 2026-05-09 | Edit this page
Estimated time: 39 minutes
Overview
Questions
- How do you write a lesson using Markdown and sandpaper?
Objectives
- Apply the CLEAR framework for effective prompting.
- Apply introspection to refine code.
- Employ prompt design patterns
- Recognise common generative AI failures
Introduction
Prompt design patterns.
Content from Explain and clean scripts
Last updated on 2026-05-09 | Edit this page
Estimated time: 33 minutes
Overview
Questions
- How do you write a lesson using Markdown and sandpaper?
Objectives
- Ask for explanations of a snippet of code to a chatbot.
- Analyse the explanations of a chatbot for understanding the snippet.
- Adapt the script to improve its clarity, modularity, and reusability.
Introduction
Something “You can’t verify what you don’t understand”
Content from Write and validate scripts
Last updated on 2026-05-09 | Edit this page
Estimated time: 34 minutes
Overview
Questions
- How do you write a lesson using Markdown and sandpaper?
Objectives
- Compare the functioning, scope, and risks of the IDE integrated LLM in relation to the chat-based LLM.
- Choose a generative AI tool for writing code according to the task.
- Write a snippet of code using generative AI directly in the IDE.
- Validate AI generated scripts using a critical framework (that the AI is doing the right thing)
Introduction
Here, we will use an integrated LLM in an IDE.
Tool selection Verification strategy Reproducibility recommendations Berkley lesson on validation
Content from Testing and refactoring scripts
Last updated on 2026-05-09 | Edit this page
Estimated time: 44 minutes
Overview
Questions
- How to safely extend the functionality of a working script?
- How do I make sure that my code is still working when its size grows?
- How do I restructure my code so that it is readable, modular and reusable?
Objectives
- Develop a suite of tests that checks the correct functioning of the generated code.
- Evaluate the quality of the tests.
- Extend the functionality of the generated code by devising a plan and using test driven development.
- Modify the structure of the generated code to improve its quality without breaking its functionality.
Introduction
Test-driven development
Content from Guiding agents from script to codebase
Last updated on 2026-05-09 | Edit this page
Estimated time: 52 minutes
Overview
Questions
- How do you write a lesson using Markdown and sandpaper?
Objectives
- Explain how agentic AI coding tools work.
- Relate identified risks of agentic methods to practical safeguards that mitigate them.
- Make use of version control by implementing Git and GitHub.
- Implement supervision strategies for AI agents in coding tasks.
- Use sandboxing and permission controls for safeguarding agent actions.
- Create a Living Spec (GEMINI.md) to guide an agent.
- Extend the functionality of the script to convert it into a structured research codebase.
Introduction
Spec-driven development. Requirements. Plan. Tests. Architecture.
Content from Reflecting and contributing
Last updated on 2026-05-09 | Edit this page
Estimated time: 12 minutes
Overview
Questions
- How do you write a lesson using Markdown and sandpaper?
Objectives
- Summarise the different generative AI tools, their scope, uses and precautions for its safe use.
- Compare your experience on the use of these tools with the experience of other users in the past.
Introduction
Reflect on what you learned and share your experience with your peers!
Content from Resources and next steps
Last updated on 2026-05-09 | Edit this page
Estimated time: 17 minutes
Overview
Questions
- What other tools are available?
Objectives
- List alternative AI coding tools.
Introduction
Design Apps Research agents