CodeOrbit: Agentic Workflow as Service (AWAS)

In Progress

A powerful, open-source platform for building and deploying agentic workflows with a user-friendly interface for creating, testing, and managing complex workflows that automate a wide range of tasks.

Node.jsExpress.jsReact.jsRedisDockerLangchainOpenAI APIFlowise
Architecture Diagram (Coming Soon)
Project Overview

CodeOrbit is a powerful, open-source platform for building and deploying agentic workflows. It provides a user-friendly interface for creating, testing, and managing complex workflows that can automate a wide range of tasks. Built on top of the popular open-source library Flowise, CodeOrbit extends its capabilities to provide a robust and scalable "Agentic Workflow as a Service" (AWAS) solution. It leverages the power of large language models (LLMs) and other AI technologies to enable developers to create sophisticated agents that can reason, plan, and execute tasks autonomously. The platform is designed with a microservices architecture, consisting of key components including the main application server, dedicated worker service, Redis for message brokering and database caching, and a React-based UI. These components are containerized using Docker and can be orchestrated using Docker Compose.

Architecture Overview

The architecture diagram above illustrates the distributed nature of the CodeOrbit platform. Key components include:

  • Main API/Workflow: Central coordination layer that manages workflows and routes requests.
  • CodeOrbit Instances: Organization-specific deployments that run within customer environments.
  • Core Libraries: Shared components including Flowise for workflow design, LangChain for agent capabilities, and LLM integrations.
  • Managed Service: Optional cloud deployment with Docker Compose for organizations that prefer not to self-host.
  • Redis Cache: Distributed cache for task queuing and state management across instances.

This architecture enables secure, cross-organizational workflows while respecting data boundaries and security requirements.

Key Features
  • Drag-and-Drop UI for building complex agentic workflows
  • Extensible and customizable platform with custom tools and integrations
  • Scalable architecture for handling concurrent workflows
  • Dockerized deployment for easy setup and deployment
  • Queue-based processing with Redis
  • Comprehensive API for programmatic access
  • Flexible deployment options (self-hosted or managed service)
  • Support for multiple LLM providers
Challenges & Solutions

Challenge:

Designing a scalable architecture for handling concurrent workflows

Solution:

Implemented a microservices architecture with Docker containerization

Challenge:

Implementing a reliable queue-based processing system

Solution:

Used Redis for reliable queue management and message brokering

Challenge:

Creating an intuitive drag-and-drop interface for complex workflows

Solution:

Built a customizable React-based UI with drag-and-drop capabilities

Challenge:

Ensuring security and isolation between different workflows

Solution:

Developed a comprehensive security model with workflow isolation

Project Info

Category

AI/ML

Duration

6 months

Team Size

3 developers

Technologies
Node.jsExpress.jsReact.jsRedisDockerLangchainOpenAI APIFlowise