RUNLOCALAIv38
->Will it run?Best GPUCompareTroubleshootStartLearnPulseModelsHardwareToolsBench
Run check
RUNLOCALAI

Independently operated catalog for local-AI hardware and software. Hand-written verdicts. Source-cited claims. Reproducible commands when we have them.

OP·Eruo Fredoline
DIR
  • Models
  • Hardware
  • Tools
  • Benchmarks
TOOLS
  • Will it run?
  • Compare hardware
  • Cost vs cloud
  • Choose my GPU
  • Prompting kits
  • Quick answers
REF
  • All buyer guides
  • Learn local AI
  • Methodology
  • Glossary
  • Errors KB
  • Trust
EDITOR
  • About
  • Author
  • How we make money
  • Editorial policy
  • Contact
LEGAL
  • Privacy
  • Terms
  • Sitemap
MAIL · MONTHLY DIGEST
Get monthly local AI changes
Monthly recap. No spam.
DISCLOSURE

Some links on this site are affiliate links (Amazon Associates and other first-class retailers). When you buy through them, we earn a small commission at no extra cost to you. Affiliate links do not influence our verdicts — there are cards we rate highly that we don't have affiliate relationships with, and cards that sell well that we refuse to recommend. Read more →

© 2026 runlocalai.coIndependently operated
RUNLOCALAI · v38
  1. >
  2. Home
  3. /Learn
  4. /Courses
  5. /Prompt Engineering Fundamentals
  6. /Ch. 8
Prompt Engineering Fundamentals

08. Personas for Different Tasks

Chapter 8 of 25 · 20 min
KEY INSIGHT

Task-specific personas produce better outputs than generic roles—define expertise, methodology, and output format for each persona.

Different tasks require different personas. A single general role rarely produces optimal output for all use cases. Matching persona to task improves relevance and reduces post-processing.

Extraction tasks need a meticulous analyst persona:

You are a data extraction specialist. Your job is to identify and extract structured information from unstructured text with high precision.

Rules:
- Extract only what is explicitly stated; do not infer
- Use null for missing fields, not placeholder text
- Preserve original formatting of dates, currencies, and names
- Flag ambiguous extractions with [AMBIGUOUS] prefix

Input: [text]
Output format: JSON

Creative tasks need a professional with clear constraints:

You are a technical writer creating user documentation. You write clear, concise sentences in active voice. You break complex processes into numbered steps. You include warnings for common mistakes.

Write an installation guide for [product].

Analysis tasks need an expert persona with methodology:

You are a financial analyst specializing in SaaS businesses. When analyzing metrics, you consider: ARR patterns, gross margin trends, net revenue retention, and customer concentration risk.

For the following financial summary, identify:
1. Metrics that indicate strong performance
2. Metrics that warrant concern
3. Questions that require additional data

Input: [financial data]

Classification tasks need a persona that understands categorization logic:

You are a support ticket classifier. You understand the difference between technical issues, billing problems, and feature requests. You look for root cause indicators, not surface keywords.

Classify this ticket and explain your reasoning:

Input: [ticket text]

Personas should include:

  • Domain expertise level
  • Decision-making criteria they apply
  • Typical output characteristics
  • Constraints on their work
{
  "role": "Code reviewer",
  "expertise": "Senior engineer, 10+ years, security focus",
  "style": "Finds specific issues, suggests concrete fixes",
  "constraints": "Prioritizes security and maintainability over convenience",
  "output": "Issue description, code location, severity, remediation"
}
EXERCISE

Create personas for three different task types in your workflow. Test each persona against five inputs and evaluate whether outputs improved.

← Chapter 7
Role Prompting
Chapter 9 →
Output Format Control