Create and deploy an LCA

Learn how to design an LCA with the AI agent, build and deploy the function, and update business configuration so Reltio invokes the LCA at the specified lifecycle hook.

Use the AI agent in the integrated development experience to create a Life Cycle Action (LCA), generate a Java project, build and deploy the function, and update business configuration so Reltio invokes it at the specified lifecycle hook.

Prerequisites

Before you begin, confirm the following requirements are met:

You completed Configure LCA development, including Maven setup, the command-line tool for your cloud, Reltio-to-cloud connectivity in Reltio Console, and collection of the environment variables required for deployment.
The integrated development experience is connected to a Reltio tenant or a Git repository that contains BusinessConfig.json or L3.reltio.json.

To create and deploy an LCA

Describe your LCA requirement to the AI agent in a natural language prompt.

Include the following details:

Detail	What to specify
Object type	The target entity or relation type.
Attributes	The attributes that the LCA reads or changes.
Lifecycle hook	When the LCA should execute, such as `beforeSave`, `afterSave`, or `validate`.
Expected behavior	Whether the LCA validates data, modifies data before save, rejects a save, or reacts after save.
Target cloud environment	AWS Lambda, Google Cloud Functions, or Azure Functions. If you don't specify one, the AI agent uses AWS Lambda.

The AI agent checks the object type, attribute, and lifecycle hook names in your request against your connected business configuration. If any name doesn't match exactly, it asks you to confirm the correct name before continuing. If the requirement only checks or flags data, the AI agent suggests a Data Validation Function (DVF) instead.

Figure 1. AI agent validating the requirement

Complete the design.
The AI agent collects any details missing from your request, helps you choose the right lifecycle hook, and confirms which attributes the LCA reads or changes. When the entity or relation type, hook, and attributes all match your business configuration, the AI agent shows the design and generates the Maven Java project in your tenant folder.
Review the generated LCA project.

The AI agent creates a Maven Java project in your tenant folder, under LifeCycle Actions/<type>-<hook>-<name>-lca/. The tenant folder is <environment>/<tenant ID>/ in your workspace, the same folder that contains Business Config and Workflows. If you work from a Git repository, the LifeCycle Actions folder is at the root of the repository instead. The project folder name is lowercase, for example LifeCycle Actions/hcp-beforesave-customerid-lca/. The project includes the LCA Java source, unit tests, a README with the business purpose, hook, and build and test instructions, and a .env.example file for deployment.

Figure 2. Generated LCA project files

To open the project, use the Explorer view in Cursor or VS Code. The project does not appear in the business configuration tree of the Reltio IDE extension view.
Prepare the local deployment settings.
The generated project folder in your tenant folder contains a file named .env.example. This file lists the environment variables required for deployment to your target cloud.
1. In the editor's file explorer, open the generated project folder.
2. Copy the .env.example file and paste it in the same folder.
3. Rename the copy to .env.
4. Open .env and enter a value for each variable listed. Use the environment variables you collected in Configure LCA development.
5. Save the file.
  
  Do not enter credential values in prompts to the AI agent or commit them to source control.

Build the LCA project.

In a terminal, open the project folder that contains pom.xml and run mvn clean package. The terminal shows one of the following results:


Result	What the terminal shows	Next step
Success	`BUILD SUCCESS`. The `target` folder contains the `.jar` or `.zip` file to deploy.	Continue to the next step to deploy the function.
Failure	`BUILD FAILURE` and the error.	Share the error with the AI agent in a prompt. The AI agent can update the generated code or tests. Then run `mvn clean package` again. For other build errors, see Troubleshoot LCA development in integrated development experience.

Deploy the function to your cloud environment.
1. The AI agent checks your .env file and names any missing values. If the project folder has no built .jar or .zip file, it asks you to build the project first. It then summarizes the deployment and asks you to confirm.
2. When the AI agent asks, approve the deployment to AWS Lambda, Google Cloud Functions, or Azure Functions. The AI agent uses the values in .env to authenticate the deployment from your workstation.
  
  Figure 3. Deployment approval in the AI agent
The AI agent reports the deployment status.
Figure 4. Successful function deployment
Review and apply the lifecycleActions update.
After a successful deploy, the AI agent proposes the change to lifecycleActions for the entity or relation type in L3.reltio.json or BusinessConfig.json. A dialog shows the configuration before and after the change. The AI agent adds only the new action, and it keeps existing actions and their order.

Figure 5. Proposed lifecycleActions update in the configuration

Select Apply to save the change to your local business configuration file. Then publish the change for your connection:
- Tenant connection: In the business configuration tree, right-click your tenant, select Apply Configuration to Tenant, and confirm the update. For more information, see Apply configuration to a Reltio tenant.
- Git repository connection: Commit and push the updated L3.reltio.json or BusinessConfig.json to your configuration repository using your standard Git workflow.
Confirm Reltio-to-cloud connectivity.

Deploying the function does not give Reltio permission to invoke it. Confirm that your cloud provider connection is configured in Reltio Console, as described in Configure LCA development. If the tenant is missing these details, the AI agent adds the Reltio Console setup steps to its summary.

Result

The LCA function is deployed to your target cloud environment, and the updated business configuration references the function through lifecycleActions. The configuration is saved to the tenant or Git repository, depending on your connection.

Verification steps

To verify the deployment, perform the following steps:

In the AI agent, confirm the deployment status shows that the function deployed successfully. You can also check the function in your cloud provider console.
Open L3.reltio.json or BusinessConfig.json and confirm that lifecycleActions on the entity or relation type lists the action under the correct lifecycle hook.
In the business configuration tree, confirm the lifecycleActions entry appears under the entity or relation type.
In Reltio Console, confirm the connection for your cloud provider is configured for the tenant.
After the configuration is applied to the tenant, save or update a test record that triggers the lifecycle hook, and confirm the LCA behaves as expected. If it does not, review the cloud provider function logs and the Reltio logs for the same time window. The AI agent can also guide you through testing the function with a sample payload.

Unify and manage your data

Create and deploy an LCA

Footer