Model steps enable integrating and executing machine learning (ML) or AI-based models or model-related operations in a Business Process (BP). You can find available Model steps in Workflow Designer of a Business Process: in the Reusable components panel, navigate to the Bot tab and find the Models category.
Bot Tasks contain work to be done by WorkFusion's bots automating human labor. Bot Tasks can exist only as a Business Process step, and you can create them using code tools or in a no-code manner.
Composite rules define a Business Process (BP) flow variant, depending on the Record column values.
Before you can use a Business Process (BP), you must link its elements by connections, thus defining the flow of data file Records across the process steps. The arrow displays the connection direction.
Unlike Bot Tasks that are completed automatically, Manual Tasks (MT) require involving a human who is either a cloud worker or an internal employee.
The no-code Data Store steps allow you to easily retrieve or modify data from Data Store tables using an SQL query. The steps send a query to a database, which then processes it and returns the result.
The no-code Rate Limiter bot step enables you to control the number of records passing through the step within a specific time frame. This is particularly beneficial for implementing rate-limiting at a granular level, such as per AI Agent, reducing the impact on resources when two AI Agents are running in a single environment.
In Work.AI, the no-code OCR step lets you accomplish the following:
A Business Process (BP) is an automation flow combining code-based and no-code elements, such as Manual Tasks, Bot Tasks, Model steps, decision or transformation rules, or nested processes. Work.AI allows you to implement two types of Business Processes (BP):
Designing a Business Process (BP) means creating and configuring a sequence of code-based or no-code steps, such as Manual or Bot Tasks, nested processes, or rules, as required to accomplish a business goal.
The guide illustrates the implementation of a schemaless Business Process (BP) for website rating from a Bot Task using the Work.AI Developer and Control Tower capabilities.
Goal
Deploy project to Control Tower
A crucial aspect of installation and further support is implementing an AI Agent so that you can change any setting outside of the implementation code without recompiling, updating, or re-deploying your AI Agent. Typically, this is achieved by making your configuration external to the source code.
The custom bot step configuration user interface is used to simplify the specification of configuration parameters for bot steps. The feature enables the creation of complex forms using frameworks like FormIO or React and is supported for Java Native Workers (JNWs), Triggers, Bot Config Bundles (BCBs), and WebHarvest tasks.
A contract defines the input and output variables in a schema-based Business Process (BP). Contracts allow designing a BP data schema in a no-code manner.
A Decision rules step allows you to configure decision rules and routing outcomes in a no-code manner.
Using the no-code Connector step, you can apply multiple configuration options to handle different connection scenarios and create custom connectors for the integration of external services into your automations.
The Transformation step is a no-code bot step that allows you to implement complex scenarios with rule conditions.
The ODF 2 provides the built-in ORM layer and comes with OrmLite as the default ORM framework used primarily to access Data Store data.
Connectors in the Work.AI platform enable Business Processes (BPs) and AI Agents to communicate with external systems. They provide standardized integration points for retrieving and enriching data, sending information to third-party services, triggering workflows based on external events, and orchestrating end-to-end automation across enterprise environments.
The article describes the legacy bot step configuration flow. For the recommended approach, refer to Configure bot steps in UI.
Business Processes (BP) with the same name have a common template, otherwise known as Definition. If you modify the design or parameters in one BP, the modification also affects all other instances with the same definition.
Templates are intended to simplify document ingestion by enabling automated information extraction from structured forms (for instance, ACORD). When a template is created, blank forms are annotated with labels. Based on these labels, data is then extracted from input documents.
Before getting into details, let's take a quick look at a simple Bot Task within the ODF 2 framework:
There are two ways to initiate the creation of a Business Process (BP):
The guide describes two automation use cases leveraging no-code Work.AI features and contract-based data schemas:
After reading this guide, you will know how to prepare and run a schemaless Business Process (BP) with a trained model and a Manual Task (MT).
After creating an ODF 2 project from the Simple Archetype, deploy it to a Control Tower instance.
In the article, the standardized Data Store schema is understood as equivalent to the ODF 2 example project's Data Store schema. With few exceptions, it utilizes concepts from Transaction and Multi-process layers of the ODF 2 architecture. If you do not use the Multi-process layer, you can ignore some entities, and this won't impact the framework integrity. In extreme scenarios, when even the Transaction layer is not used, the Data Store schema should be designed from scratch.
Data flow
The article dwells on the structure of a project created from the ODF 2 Simple Archetype. For more details on creating a new project, see Start Maven project.
What is Assembly?
Maven archetypes are templates that let you quickly generate projects from a predefined set of core files. The guide describes how to use the ODF 2 Simple Archetype to generate a new project.
Control Tower (CT) is the central orchestration application of the Work.AI platform. It enables users to perform the following actions:
The qualification mechanism is no longer available for access control purposes and is replaced by a new user management approach. However, to support training before employees begin performing actual tasks, WorkFusion provides a special Qualification Business Process (BP), which you can deploy and configure as described below.
With the Asset Bundle import feature, everything required for a Business Process (BP) delivery is now conveniently wrapped in a self-sufficient package, which can be ported to a target environment in few simple steps. The import mechanism supports large files and is less error-prone than traditional methods. Successful import is asynchronous—meaning fast—and verified by a checksum. This accelerates BP delivery, decreases the total cost of development, and eventually contributes to continuous integration.
The article lists operators supported for no-code Transformation and Decision rules steps.
Design concepts
With ODF 2, we encourage you to use Data Stores as a universal and effective way to have different steps of Business Processes within your AI Agent exchange data. The section provides a reference on how to organize your Data Model through AI Agent automation.
The guide explains how to add a Manual Task to a Business Process based on ODF 2. When working with Manual Tasks, consider the following aspects:
You can create a package with a Business Process (BP) and import it from one WorkFusion environment to another, for example, from DEV to UAT or PROD.
ODF 1 is not supported in Work.AI v10.3, and you are encouraged to fully transition to ODF 2. For the ODF documentation, refer to the v10.2.9 user guide.
ODF 1 is not supported in Work.AI v10.3, and you are encouraged to fully transition to ODF 2. For the ODF documentation, refer to the v10.2.9 user guide.
The instruction describes running, stopping, and scheduling Business Processes based on the ODF 2 example project.
The Orchestration section guides you through the steps to deliver and orchestrate automations, including the involvement of a human worker in the loop. The figure below illustrates the actions available for the purpose in the Control Tower (CT) and Workspace (WS) user interfaces.
The article describes an AI Agent Asset Bundle. For details on the structure of an AI Agent Variation Asset Bundle, refer to Package assets into AI Agent Variation Asset Bundle.
The article describes an AI Agent Variation Asset Bundle.
Stateless execution definition
To quickstart with Spoke, watch this tutorial video and practice the creation of a basic Spoke test.
There are some useful prepackaged task types not associated with any APIs. You can use them in the standard cross-process flow.
The REST (HTTP) connector is an external connector that listens to incoming HTTP requests and sends records to Control Tower (CT). Optionally, for sync requests only, it can also return transaction results as an HTTP response.
The ODF 2 example project is created to demonstrate key features of the framework. It contains four simple Business Processes that operate on a mock data model. While functional, these processes do not perform anything practically useful.
ODF 2 allows splitting a Transaction into multiple sub-transactions and joining them back. This is usually done to process some data from the original Transaction in parallel.
Split and merge records
If you have several Business Processes (BP) where one BP produces data for another, you can apply the event-based communication mechanism and configure the data transfer between them. You can set a sub-process to start from another BP based on a specific event or set transaction processing to be launched from another transaction in a different BP using no-code sub-process steps.
You can start a Business Process (BP) from a Manual Task. Below are some code samples of how you can perform various operations with it.
To start a Business Process (BP) based on external events or scheduled polling of a source system, you can apply the no-code Trigger element.
The article is an example of the OCR Bridge step usage in the scope of the ODF 2 framework. The Bot Tasks used in the example are a part of Processing Business Process from the example project.
Why WorkFusion recommends using version control
To access the list of Business Processes (BP), log in to Control Tower and go to Business Processes > View All.
The article is a tutorial on how to create an end-to-end Business Process (BP) for information extraction (IE), comprising the following steps:
A commonly encountered task in ODF 2 is waiting for the completion of all transactions in a given Business Process. There is no out-of-the-box solution that fits all use cases. For example, a Monitor Task often serves as the source of transactions, making it impossible to know when new ones appear. However, once all requirements are known, such a feature is quite easy to implement.
To illustrate how user Transactions are created and processed, let's simulate a simple scenario for a Business Process (BP) consisting of two Bot Tasks: