AI Lesson & Submodule
Instruction Design
Deconstruct system instructions, delimiters, and target response structures.
Why This Matters
Clear directives and delimiters isolate untrusted user inputs, preventing parsing errors.
What You Will Learn
- •Use text delimiters
- •Write clear constraints
- •Optimize instruction paths
Concepts Covered
XML DelimitersPIN rulesDirectives layouts
Mapped Foundation Project: AI Scam Detector
System design analyzing incoming messages for fraud signals, using classification prompt instructions and confidence bounds.
Architecture Preview
Inbound message router evaluating inputs against prompt structures and returning classified risk vectors.
User Input TextScam Evaluator PromptScam Model Call
Tech Stack Planned
ReactTypeScript
GitHub: Coming SoonLive Demo: Coming Soon
Coming SoonTechnical Interview Value
- ?Explain how XML delimiters sandbox user inputs from model system rules