Abstraction Levels and User Personas
You'll learn how to design platform abstractions that match different user skill levels and complexity needs, ensuring both power users and beginners can work effectively.
The Core Idea
Not all platform users are the same. A junior developer deploying their first web app needs simple, opinionated interfaces. A senior platform engineer troubleshooting a complex distributed system needs deep access to underlying configuration. Effective Internal Developer Platforms (IDPs) offer multiple abstraction levels tailored to distinct user personas, like a car that offers both automatic and manual transmission modes.
Why Personas Matter
Just as "golden paths" provide opinionated defaults, user personas help you understand who needs what level of abstraction:
- Application developers may want: "Deploy my app" with minimal infrastructure decisions
- Platform engineers may need: Full control over networking, security policies, and resource configuration
- Security teams might require: Policy enforcement points and audit capabilities
Creating these layers prevents the platform from becoming either too restrictive (frustrating advanced users) or too complex (overwhelming beginners).
Designing Abstraction Layers
Think of abstraction levels as an elevator in a building:
- Top floor (highest abstraction): Simple CLI commands or UI forms — "deploy app, choose region"
- Middle floors: Template-based deployments with some customization options
- Ground floor (lowest abstraction): Direct infrastructure access via IaC for maximum flexibility
Each persona enters at their appropriate floor. The platform provides escape hatches so users can move down when they outgrow simpler abstractions, but defaults guide them to appropriate levels.
Balancing Act
The challenge isn't building every possible abstraction — it's identifying your actual user personas, understanding their workflows, and creating just enough layers to serve them without fragmenting your platform into unmaintainable complexity.
Key Takeaway: Match platform abstractions to real user personas and skill levels — provide simple defaults for common cases while offering deeper access for complex scenarios, creating a scalable self-service experience.