Role Prompting: Making AI Your Expert

What Is Role Prompting?

Role prompting means assigning a specific expertise role to the AI before giving your task. Instead of "Write an email," you say "You are a senior business communication specialist. Write an email."

Why It Works

Language models are trained on vast, diverse data. When you assign a role, you narrow the model's focus to the most relevant subset of its training — the way an expert in that field would write.

Effective Role Examples

• "You are a Nobel Prize-winning economist..."
• "You are a senior UX designer with 15 years of experience..."
• "You are a Pulitzer Prize-winning journalist..."
• "You are a Harvard law professor..."

Making Roles More Effective

1. Be specific: "Senior Python developer" beats "developer"
2. Add experience level: "15 years of experience" adds authority
3. Include domain: "B2B SaaS marketing" beats "marketing"
4. Combine roles: "You are both a data scientist and a science communicator..."

Common Mistakes

• Assigning unrealistic roles ("You are the world's best...")
• Not matching the role to the task
• Using roles that don't change the output meaningfully

Try It

Browse our prompts to see role prompting in action across many scenarios.