Beginner LevelAI, LLMs, Embeddings & MCP
What is the difference between System Prompts and User Prompts in LLM API calls?
❓ Interview Prompt: Explain message role hierarchies (system, user, assistant, tool) in chat completion APIs.
Modern LLM Chat APIs structure interactions using message arrays with explicit roles:
1. System Role (system):
Sets the overarching behavioral persona, constraints, output formatting rules, and safety boundaries. Given higher priority in model instruction following.
2. User Role (user):
Contains the input query, message payload, or question submitted by the end user.
3. Assistant Role (assistant):
Contains previous responses generated by the LLM in a multi-turn conversation.
4. Tool Role (tool):
Contains results returned by external API function executions back to the model.
Code Implementation Example
typescript
const messages = [
{ role: 'system', content: 'You are a concise TypeScript code reviewer. Respond ONLY with code diffs.' },
{ role: 'user', content: 'How do I check if an array is empty in JS?' },
{ role: 'assistant', content: 'arr.length === 0' },
{ role: 'user', content: 'What about null checking?' }
];💡Multi-turn conversation payload structuring system, user, and assistant roles.
🎯 Interview Key Takeaway
System prompts establish model persona and rules; user prompts supply dynamic runtime queries.
#System Prompts#User Prompts#Chat API#Beginner
Back to all AI, LLMs, Embeddings & MCP Questions→