ProcessOS

AI model by Camunda

The agentic operating system for your business processes.

ProcessOS is the agentic intelligence layer for the Camunda platform that discovers, re-engineers, and deploys AI-native business processes. It enables enterprises to transform legacy workflows into outcome-based systems in weeks rather than quarters using specialized AI agents and organizational memory.

Model specs

The full spec card (context, pricing per token, modalities and more) on AntSpace. Sign up free to see the full spec card.

Launched
2026-05-15

Category: Workflow Automation

Released
May 2026 (announced at CamundaCon)
Discovery Timeline
1–2 weeks
Re-engineering Timeline
1 week
Build and Deploy Timeline
2–4 weeks
Primary Interface
Claude Code (Coding Agent)
Storage Integration
Project Repository
  • Discovery in 1–2 weeks instead of months
  • Build and deploy in 2–4 weeks instead of quarters
  • Agentic interface using Claude Code integration
  • Outcome-based redesign using natural language and KPIs
  • Organizational memory for faster subsequent transformations

Overview

ProcessOS is the intelligence layer of the Camunda platform designed to help enterprises re-engineer business processes for an AI-native world. It utilizes specialized AI agents to discover existing workflows, redesign them around desired outcomes, and accelerate the build and deployment phases. The platform moves beyond simple task automation by enabling structural redesign that combines AI reasoning, human judgment, and deterministic systems. Developers and architects interact with ProcessOS through coding agents like Claude Code to write to a shared project repository. ProcessOS compresses traditional transformation timelines, moving from kickoff to production in weeks rather than quarters.

Process Lifecycle Intelligence

  • Process Discovery: Specialized AI agents analyze documentation, process models, and operational data to discover how work actually runs.
  • Outcome-Based Redesign: The system reconstructs as-is processes and designs to-be processes based on natural language descriptions and defined KPIs.
  • Rapid Build and Deploy: ProcessOS generates process models, integrations, data mappings, agent prompts, decisions, and UI forms to move designs toward deployment.
  • Organizational memory captures decisions, systems, and integration patterns to make each subsequent process transformation faster.

Developer Experience

  • Agentic Interface: Architects and developers interact with the platform using coding agents such as Claude Code.
  • Project Repository Integration: The system writes directly to a shared Project Repository to maintain a single source of truth for process logic.
  • ProcessOS Agent Skills: ProcessOS uses Agent Skills to allow developers to work with the intelligence layer.

Optimization and Governance

  • KPI-Driven Optimization: The platform builds a weighted scoring model from KPIs such as cycle time, manual effort, quality, cost, and throughput.
  • Continuous Improvement: ProcessOS evaluates live process performance against defined outcomes and tests proposed changes against historical data.
  • AI-Native Re-engineering: The system focuses on structural change by redesigning processes around what AI, people, and systems should each do.
  • Extensible Plugin Model: An extensible plugin model allows process mining tools like Celonis or MPMX to enrich the operational picture.

Use cases

  • Enterprises use ProcessOS to discover how legacy work actually runs by analyzing documentation and operational data.
  • Architects redesign customer onboarding processes from months to days by focusing on outcomes rather than existing manual steps.
  • Developers use Claude Code to interact with ProcessOS Agent Skills and write process logic directly to a shared repository.
  • Operations teams continuously improve live processes by testing AI-proposed changes against historical data before production deployment.
  • Organizations build organizational memory of integration patterns and decisions to accelerate subsequent process transformation projects.

Specifications

Access
Intelligence layer for the Camunda platform accessed via coding agents like Claude Code
Released
May 2026 (announced at CamundaCon)
Discovery Timeline
1–2 weeks
Re-engineering Timeline
1 week
Build and Deploy Timeline
2–4 weeks
Primary Interface
Claude Code (Coding Agent)
Storage Integration
Project Repository
KPI Metrics
Cycle time, Manual effort, Quality, Cost, Throughput, Compliance
Supported Plugins
Celonis, MPMX

How it works

  1. Define Outcomes and KPIs: Users describe the desired process in natural language and define the KPIs that matter, such as cycle time or cost.
  2. Discover and Re-engineer: Specialized AI agents analyze existing documentation and data to reconstruct the current process and design the new AI-native version.
  3. Build and Deploy: The system generates process models, integrations, and UI forms, writing them to a Project Repository for deployment on the Camunda platform.
  4. Continuous Improvement: ProcessOS monitors live production data against KPIs and tests proposed improvements against historical data before human review.

FAQ

How fast can ProcessOS deploy new processes?

Traditional transformation can take months for discovery and quarters for deployment. ProcessOS compresses discovery to 1–2 weeks, re-engineering to 1 week, and build/deploy to 2–4 weeks.

How do developers interact with ProcessOS?

Developers and architects interact with ProcessOS through coding agents like Claude Code, which use ProcessOS Agent Skills to write to a shared Project Repository.

What is organizational memory in ProcessOS?

ProcessOS captures decisions, systems, integration patterns, and edge cases in an organizational memory. This allows each subsequent process transformation to move faster by retaining organizational context.

Added by AntSpace from public sources (camunda.com).

Related

More AI models by Camunda