Prodigy - OpenText Developer Admin - IM Developer Administration Integration and Automation

Common Integration Use Cases Between Prodigy and OpenText Developer Admin - IM Developer Administration

Below are practical integration scenarios where Prodigy and OpenText Developer Admin - IM Developer Administration can work together to support AI data operations, controlled environment management, and enterprise integration governance.

1. Controlled ingestion of enterprise data into Prodigy for model training

Flow: OpenText Developer Admin - IM Developer Administration to Prodigy

Integration teams can use OpenText to manage secure API access, credentials, and messaging routes that deliver approved source data such as documents, images, or transaction records into Prodigy for annotation. This is useful when training AI models on enterprise content stored in trading grid or integration pipelines.

• Ensures only authorized datasets are exposed to labeling teams
• Separates development and production data sources
• Reduces manual file handling and data leakage risk

2. Automated export of labeled datasets back into enterprise integration pipelines

Flow: Prodigy to OpenText Developer Admin - IM Developer Administration

Once data is labeled in Prodigy, the annotated output can be pushed through OpenText-managed integration endpoints to downstream systems such as data lakes, model training services, or document processing workflows. This supports repeatable handoff from annotation to operational AI use.

• Speeds up model training cycles
• Standardizes dataset delivery to downstream consumers
• Supports auditability of labeled data movement

3. Environment-specific API and credential governance for annotation workflows

Flow: Bi-directional

OpenText Developer Admin can manage separate development, test, and production credentials for Prodigy integrations, while Prodigy can be configured to use the correct endpoints for each environment. This is valuable for teams validating annotation workflows before promoting them to production AI pipelines.

• Prevents accidental use of production credentials in test environments
• Supports controlled rollout of new labeling projects
• Improves compliance with enterprise change management practices

4. Event-driven labeling queue creation from integration messages

Flow: OpenText Developer Admin - IM Developer Administration to Prodigy

When OpenText receives new business events such as customer complaints, claims, or quality inspection records, it can route selected items into Prodigy as labeling tasks. This enables domain experts to annotate high-value cases as they arrive, rather than waiting for batch exports.

• Supports near real-time AI training data creation
• Prioritizes the most relevant records for active learning
• Improves responsiveness for operational AI use cases

5. Governance of annotation-related integration artifacts and scripts

Flow: OpenText Developer Admin - IM Developer Administration to Prodigy

Integration artifacts such as API definitions, transformation scripts, and message mappings used to move data into and out of Prodigy can be centrally managed in OpenText. This helps AI and integration teams maintain version control and consistent deployment practices across labeling projects.

• Creates a single control point for integration logic
• Reduces configuration drift between teams
• Improves maintainability of custom annotation pipelines

6. Feedback loop from model errors to new labeling tasks

Flow: Prodigy to OpenText Developer Admin - IM Developer Administration to Prodigy

When a deployed model produces low-confidence predictions or errors, those records can be sent through OpenText-managed messaging back into Prodigy for relabeling. This creates a closed-loop process for continuous model improvement using real production exceptions.

• Turns production failures into training opportunities
• Supports active learning and targeted retraining
• Helps data science teams focus on the most impactful examples

7. Secure collaboration between data science and integration teams

Flow: Bi-directional

Data scientists working in Prodigy and integration developers working in OpenText can collaborate through shared integration endpoints, approved data contracts, and controlled access policies. This is especially useful in enterprises where AI initiatives depend on integration teams to expose source systems safely.

• Aligns annotation workflows with enterprise integration standards
• Improves handoff between AI and middleware teams
• Supports scalable cross-functional delivery for AI projects

These integration patterns help enterprises connect AI annotation work in Prodigy with the governance, messaging, and environment control capabilities of OpenText Developer Admin - IM Developer Administration, enabling more secure and efficient AI delivery pipelines.