Back to projects
Multi-agent System
AI Agents
A multi-agent AI system orchestrated with LangGraph: a router classifies intent and dispatches to specialized agents (CRM and finances) that act by connecting to the platform services. Powered by OpenAI and Anthropic models, exposed as a FastAPI service.
PythonFastAPILangGraphLangChainOpenAIAnthropicDocker
Some details omitted under NDA
2
Specialized agents
2
AI models
22
Test files
What the system does
Intent routing
A router classifies intent and dispatches to the right agent.
Specialized agents
A CRM agent and a finances agent, each with its own scope.
Integrated actions
Connects to platform services (CRM, finances) to act.
Multi-model
Uses OpenAI and Anthropic models.
LangGraph orchestration
Stateful node graph for reliable agent flows.
The engineering behind it
- Agent orchestration with LangGraph (graph + intent router).
- Multi-model: OpenAI and Anthropic.
- Containerized FastAPI service (Docker).
- Integration with CRM and finances services for real actions.