CODESEEDSign in
Explore projects
Pythonadvancedai

Multi-agent Research Pipeline

Automated research pipeline with specialized agents: an orchestrator distributes sub-questions, agents search and summarize, results are aggregated.

5 steps

Project steps

  1. 01

    Orchestrator Agent

    Receives a complex question, breaks it down into 3-5 sub-questions with an LLM call.

  2. 02

    Search Agent

    Each sub-question goes to an agent that searches with DuckDuckGo and extracts relevant content.

  3. 03

    Summarizer Agent

    Summarizes the content of each search result into ~200 tokens.

  4. 04

    Aggregator

    Combines all summaries and generates a coherent final answer with citations.

  5. 05

    Streaming Output

    Streams the final answer progressively with Server-Sent Events.

Recommended resources

Ready to build this?

Fork the repo on GitHub and start building. A mentor will review your code when you open a PR.

5 steps

Tech stack

Pythonasyncioanthropicduckduckgo-searchpydantic