Geist is a process-intelligence platform for understanding how an organization works, improving that operating model, and keeping the model connected to real operations.
The operating loop
Geist organizes work as a continuous loop:
- Model processes and their intended behavior.
- Connect context such as owners, systems, locations, suppliers, controls, and dependencies.
- Observe operational events and external signals that may affect the process.
- Analyze and improve the model using evidence, simulation, and review.
- Apply changes with control, then observe the new state.
The process model remains the central reference. Enterprise context explains where and why the process operates, while observed data shows what is happening.
Core product surfaces
Process Modeler
The Process Modeler is the BPMN workspace. Teams can create, inspect, review, and improve process models while keeping changes visible and auditable.
Process Finder
Process Finder organizes process assets and related evidence so teams can locate models, compare variants, and understand their status.
Enterprise Map
The Enterprise Map connects processes with the organization around them: responsibilities, systems, locations, flows, dependencies, and relevant external signals.
AI agents
Geist agents use process and enterprise context to investigate questions, prepare changes, and help teams carry out reviewed work. Available actions remain subject to the user’s permissions and write mode.
How the surfaces work together
A process model can identify a task, but the surrounding context answers the operational questions:
- Who owns it?
- Which system supports it?
- Where does it happen?
- Which upstream or downstream flows depend on it?
- Which signals indicate risk or change?
Geist keeps those answers connected instead of treating them as separate diagrams and spreadsheets. This provides a shared basis for analysis, simulation, implementation, and later learning.
Choose your next step
Use the product guides to understand a specific surface. Customer operating instructions, deployment material, and implementation references remain outside this public library until their classification and access model are approved.
Public documentation explains product concepts and approved capabilities. It does not expose organization data, internal runbooks, or unreviewed API operations.