How To Instantly 10x Your AI Prompts

Hi,

Welcome to this week's edition of Leverage AI.

Today, I’m excited to share a simple trick you can use to instantly 10x your AI prompts.

You'll learn:

• How XML tags bring clarity and structure to your prompts

• Why this technique can lead to more accurate and maintainable AI responses

• Practical examples to help you get started

Let's dive in.

The Problem With Unstructured Text

Ever find your AI prompts turning into a jumbled mess? It happens surprisingly often. Complex prompts can spiral into confusion and become unreliable. For ad-hoc and one-off prompts, you can get away with unstructured text, but for reliable and easily maintainable prompts, that approach won’t scale.

Without a clear structure, your prompts become harder to edit, sections get mixed up, and LLMs have difficulty knowing where one section starts and another ends.

Here's a straightforward fix: use XML tags to organise your prompts.

Think of XML tags as simple containers that keep different parts of your prompt neatly separated. They're like clear signposts that you and the AI can easily follow. Since AI models are particularly good at reading XML, this approach brings some practical benefits:

• You can see exactly where each section starts and ends

• Need to change something? Just update that specific section

• Fewer confusing responses because your instructions are crystal clear

And don't worry – I'm not suggesting you write large verbose XML code everywhere. Just use basic tags to keep your prompts organised.

Accuracy

When you use XML tags, the AI understands precisely what you want because each piece of information has its place. Instead of wading through a wall of text, the AI can quickly identify roles, tasks, and data.

Here's what it looks like in practice:

You will be provided with a pair of articles about the same topic. 

First, summarize the arguments of each article. 

Then indicate which of them makes a better argument and explain why.

<article>Insert first article here </article>

<article>Insert second article here </article>

Using delimiters might not make a difference for straightforward tasks such as these. However, the more complex a task is, the more critical it is to disambiguate task details. Don’t make the model work to understand your prompt.

Readability

XML tags don’t just help the AI - they make your prompts easier for you to read and troubleshoot. When everything has its place, you can quickly scan and find exactly what you need to modify.

Common tags you might use:

• <data> for information the AI needs to work with

• <output> for output formatting

• <examples> for providing examples

The best part? This approach works across all AI models - whether you're using OpenAI, Claude, or Gemini. The structured approach remains effective because it's based on clear organisation, not platform-specific tricks.

Further Reading

Want to dive deeper? Check out these resources:

• Anthropic's Guide on Using XML Tags

• OpenAI's Guide on Prompt Engineering

• Google on Prompt Structuring

💡 Wrap Up

Key Takeaways:

1. Structure makes prompts clearer and easier to maintain

2. XML tags will improve your prompts

3. You don't need complex XML - basic tags are enough

The real value isn't in the XML itself - it's in having a clear system for organising your prompts that both you and the AI can easily follow.

Thanks,

Owain

Your Opinion Matters

What did you think of today’s email? Your feedback helps me create better emails for you!

Got more feedback or want me to cover a specific topic? Reply to this email and let me know.

Owain