KiwiQ
StaleDescription
Production-grade multi-agent orchestration platform with JSON-defined agents, multi-tier memory, and built-in observability, battle-tested on 200+ enterprise AI agents with full enterprise deployment support.
Key Features
- SDK-first workflow engine with Python/JSON graph schemas compiled to LangGraph and executed via Prefect
- Multi-provider LLM support including OpenAI, Anthropic, Google Gemini, Perplexity, and AWS Bedrock
- Human-in-the-loop with real-time WebSocket streaming for workflow pause, review, and approval
- 24+ reusable node types for LLM calls, routing, conditional branching, data transforms, and code execution
- Multi-tier memory architecture: PostgreSQL for state, MongoDB for documents, Weaviate for vectors, Redis for cache
- Production-ready with 27+ workflow definitions for content creation, diagnostics, lead scoring, and deep research
Use Cases
💡 Enterprise multi-agent orchestration with complex multi-step AI workflows
💡 Marketing automation using pre-built content creation and lead scoring workflows
💡 RAG pipelines with document ingestion, vector search, and retrieval-augmented generation
💡 Deep research workflows combining web scraping, search, and multi-source data synthesis
💡 Customer data management with versioned document storage and CRUD workflow nodes
Categories
Quick Start
Clone the repository, install dependencies with Poetry, configure environment variables in .env, then start all services with `docker compose -f docker-compose-dev.yml up --build`. Access the API docs at localhost:8000/docs and the Prefect dashboard at localhost:4201.