Building a Multi-Agent Research System: Architecture and Lessons

Why Multi-Agent for Research?

A single LLM context cannot simultaneously hold search results, source analysis, cross-source comparisons, and synthesis conclusions. Multi-agent systems break this into parallel specialized workstreams.

Architecture

1. Orchestrator: decomposes research question, assigns to specialists, synthesizes results
2. Specialist Agents: web search, document analysis, data extraction, fact-checking
3. Synthesis Agent: combines outputs into final report

def decompose_question(main_question: str) -> list:
    import json
    response = client.messages.create(
        model='claude-opus-4-6', max_tokens=1024,
        messages=[{'role': 'user', 'content': f'Break into 3-5 focused sub-questions:\n{main_question}\n\nReturn as JSON list.'}]
    )
    return json.loads(response.content[0].text)

Production Lessons

• Minimize agent handoffs -- each adds latency
• Synthesis agent must detect and resolve conflicting information from specialists
• Use Haiku for lightweight tasks, Opus only for final synthesis
• Compress results before inter-agent handoffs to control context size