Skip to content

5. Orchestrator-Worker

Mini-Project: Competitive Intelligence Report Generator

An orchestrator that dynamically plans and dispatches research workers to gather and synthesize competitive intelligence on companies.

View on GitHub


Description

In the Orchestrator-Worker pattern, a central orchestrator LLM dynamically breaks down a task into subtasks, delegates them to worker LLMs, and synthesizes their results. Unlike Parallelization (where subtasks are pre-defined), the orchestrator dynamically determines what subtasks are needed based on the specific input.

When to Use

  • Complex tasks where subtasks cannot be predicted in advance
  • Coding tasks that require changes across multiple files
  • Research tasks that need adaptive exploration strategies
  • When the decomposition itself requires intelligence

Benefits

Benefit Description
Adaptability Subtask decomposition is dynamic, not hardcoded
Scalability Can spawn as many workers as needed
Intelligence Orchestrator applies reasoning to task breakdown
Synthesis Final output integrates all worker contributions

Architecture Diagram

flowchart TD
    A[User Input] --> B[Orchestrator LLM]
    B -->|Plan & Delegate| C[Worker 1]
    B -->|Plan & Delegate| D[Worker 2]
    B -->|Plan & Delegate| E[Worker N]
    C --> F[Results]
    D --> F
    E --> F
    F --> B
    B -->|Synthesize| G[Final Output]

    style A fill:#4CAF50,color:#fff
    style B fill:#FF5722,color:#fff
    style C fill:#2196F3,color:#fff
    style D fill:#9C27B0,color:#fff
    style E fill:#00BCD4,color:#fff
    style F fill:#FFC107,color:#000
    style G fill:#4CAF50,color:#fff