AI Agent
Build conversational agents with persistent memory, tool calling capabilities, and structured decision-making workflows.
Architecture Patterns
Architecture patterns provide proven solutions to common application challenges using Raindrop framework components. Each pattern combines multiple components to solve specific problems with clear guidance on implementation and best practices.
Available Patterns
API Gateway
Create unified API endpoints that route requests to multiple backend services with authentication, rate limiting, and monitoring.
Background Job Processing
Process long-running tasks asynchronously with reliable queue-based architecture and automatic retry mechanisms.
Multi-Agentic System
Coordinate multiple AI agents working together on complex tasks with shared state management and inter-agent communication.
Retrieval Augmented Generation
Enhance AI responses with relevant context from document collections using semantic search and intelligent retrieval.
Using Architecture Patterns
Each pattern includes:
- Overview - Problem solved and when to use the pattern
- Architecture Diagram - Visual representation of component relationships
- Components - List of required Raindrop framework resources
- Logical Flow - Step-by-step process breakdown
- Implementation - High-level setup and deployment guidance
- Best Practices - Key recommendations and common pitfalls
Pattern Structure
Architecture patterns are designed for reusability and combine multiple Raindrop components:
- Actors - Stateful compute with persistent data
- Services - HTTP request handlers
- SmartMemory - AI agent memory systems
- SmartBucket - Document storage with semantic search
- AI Models - Language models and embeddings
- Queues - Asynchronous message processing
- Observers - Event-driven reactive components
Choose patterns based on your application requirements and combine them as needed for complex systems.