System Prompt
What is a system prompt?
A system prompt is the foundational instruction set given to an LLM that defines its role, behavior, tone, constraints, and capabilities for a particular use case or application. Unlike user prompts—which change with each interaction—the system prompt remains constant and establishes the context within which all user interactions occur.
For AI product builders, the system prompt is the primary way to configure how an AI assistant behaves and responds.
What goes into a system prompt?
System prompts typically include several key elements:
Role and purpose. The system prompt defines what the AI is supposed to do, such as "You are a customer interview coach who helps product managers improve their interviewing skills."
Behavioral guidelines. Teams specify tone, constraints, and how the AI should handle different scenarios. This might include instructions about when to ask clarifying questions, how to structure responses, or what topics to avoid.
Evaluation criteria. System prompts can include rubrics or frameworks that guide how the AI should assess inputs and generate outputs, ensuring consistency across interactions.
Context and domain knowledge. Teams often combine system prompts with uploaded files, project rules, or domain-specific information that the AI needs to reference.
How is a system prompt different from a user prompt?
The system prompt is persistent—it stays the same across all interactions and defines the AI's fundamental behavior. User prompts, in contrast, are dynamic and change with each interaction. They represent what the user is asking in that specific moment.
Think of the system prompt as the AI's "job description" and user prompts as the specific tasks users ask it to perform within that role.
Learn more:
- Behind the Scenes: Building the Product Talk Interview Coach
- How I Designed & Implemented Evals for Product Talk's Interview Coach
Related terms: