Weekly Digest
Chain of Thought Takes Center Stage
Jan 8 – Jan 14, 2024
Reasoning techniques dominated this week. The step-by-step prompting tip gained traction, alongside structured approaches for complex coding tasks.
Top Tips This Week
All Tips from This Week
How do I get Claude to write in my style?
Paste 3-5 examples of your writing and ask Claude to analyze the style before writing new content. The model will identify patterns in tone, sentence structure, and vocabulary that it can then replicate.
How should I structure complex coding prompts?
Break complex coding tasks into three parts: context (what exists), goal (what you want), and constraints (requirements and limitations). This structure prevents the AI from making assumptions and produces more accurate code.
Why should I ask the AI to think step by step?
Adding "think step by step" or "let's work through this" to prompts triggers chain-of-thought reasoning. This dramatically improves accuracy on math, logic, and multi-step problems by forcing the model to show its work.
How do I use AI to refactor legacy code safely?
Feed the AI your existing tests first, then the code to refactor. Ask it to preserve all test-passing behavior while improving structure. This anchors the refactoring to verified behavior rather than assumptions.
What is the "sandwich" feedback technique?
When iterating on AI output, structure feedback as: what worked, what to change, then restate what to keep. This prevents the model from over-correcting and throwing out good elements while fixing problems.
How can I get better code suggestions in Copilot?
Write a detailed comment describing the function before you start typing code. Copilot reads comments as context, so a clear description of inputs, outputs, and edge cases produces dramatically better suggestions.
What is context stuffing and when should I use it?
Paste relevant documentation, examples, or reference material directly into your prompt before asking questions. AI models work best with explicit context rather than relying on training data that may be outdated or incomplete.