A multinational bank is designing an AI compliance platform using the Anthropic Claude API.
The system uses multiple specialized agents:
• one agent analyzes transaction anomalies,
• another reviews regulatory obligations,
• and a third drafts escalation summaries.
During testing, the platform produces inconsistent final reports because agents independently generate conclusions without coordination.
Which orchestration strategy BEST improves consistency across the multi-agent workflow?

Question

Accepted Answer

Introduce a coordinator agent that aggregates outputs, resolves conflicts, and produces the final response.

Questions

Related questions