Domain-Specific Prompts
Coding, writing, analysis, data — each domain has its own prompting patterns. Master them all.
What You'll Learn
- Prompting patterns for software development
- Prompting patterns for writing and content creation
- Prompting patterns for data analysis and research
- How to adapt your general skills to any domain
Prompts for Software Development
Coding prompts need precision. The AI must understand your stack, your patterns, and your constraints — or you'll spend more time fixing its code than writing your own.
Code Generation
"Write a TypeScript function that validates email addresses. Requirements: RFC 5322 compliant, returns { valid: boolean, reason?: string }, handles edge cases (plus addressing, international domains). Use no external libraries. Include JSDoc comments. Write 3 unit tests using Vitest."
Code Review
"Review this function for: security vulnerabilities, performance issues, error handling gaps, and readability. For each issue found, explain the risk, show the problematic line, and provide a fix. If the code is solid, say so. Don't nitpick style preferences."
Debugging
"This function returns undefined when the input array has duplicate values. Here's the function: [code]. Here's a failing test case: [test]. Walk through the execution step by step with the failing input. Identify exactly where the logic breaks and why."
Prompts for Content Creation
Writing prompts need voice and audience clarity. Without them, you get perfectly grammatical content that sounds like it was written by nobody, for nobody.
Blog Post
"Write a blog post about [topic]. Audience: [who]. They already know [baseline knowledge] but don't know [what this teaches]. Tone: conversational, like explaining to a smart friend over coffee. Use concrete examples, not abstract theory. 800-1000 words. Include a practical takeaway they can use today."
Email Sequence
"Write a 3-email welcome sequence for new subscribers to [product]. Email 1 (Day 0): Welcome + immediate value. Email 2 (Day 3): The #1 mistake beginners make. Email 3 (Day 7): Soft pitch for [paid offering]. Each email: subject line, preview text, body under 200 words. Tone: helpful mentor, not salesperson."
Prompts for Data and Research
Analysis prompts need structure and rigor. You want the AI to think systematically, not just generate plausible-sounding conclusions.
Data Analysis
"Analyze this sales data. First: identify the top 3 trends. For each trend, quantify the change (% or absolute), identify the likely cause, and rate your confidence (high/medium/low). Then: flag any anomalies that don't fit the trends. Finally: recommend 2 actions based on your analysis."
Research Synthesis
"I have 5 articles about [topic]. For each, extract: main argument, key evidence, methodology used, and limitations. Then synthesize across all 5: where do they agree? Where do they contradict? What questions remain unanswered? Present as a research brief."
Adapting to Any Domain
Every domain has the same underlying needs. When entering a new domain, ask yourself:
What does "good" look like here? Define quality criteria specific to this domain.
What are the common mistakes? Tell the AI to avoid domain-specific pitfalls.
What terminology matters? Use the right jargon so the AI activates the right knowledge.
What's the expected output format? Every domain has conventions. Match them.
This lesson is for Pro members
Unlock all 520+ lessons across 52 courses with Academy Pro.
Already a member? Sign in to access your lessons.