Prodigy - OpenText Core Case Integration and Automation

Common Integration Use Cases Between Prodigy and OpenText Core Case

1. Case-Driven Data Labeling for Model Training

Flow: OpenText Core Case ? Prodigy

When a case requires AI support, such as claims triage, complaint classification, or compliance review, OpenText Core Case can send relevant case documents, notes, emails, and attachments to Prodigy for annotation. Data scientists and subject matter experts can label the content to create training data for text classification, entity extraction, or document routing models.

Business value: Speeds up model development using real operational cases, improves label quality with context-rich case data, and helps automate repetitive case handling tasks.

2. Human Review and Exception Handling for AI Predictions

Flow: Prodigy ? OpenText Core Case

Predictions from machine learning models trained or refined in Prodigy can be pushed into OpenText Core Case when confidence is low or when a case needs human review. For example, a model may flag a document as potentially fraudulent, non-compliant, or misrouted, and OpenText Core Case can create a review case for investigation and resolution.

Business value: Creates a controlled human-in-the-loop process, improves decision accountability, and ensures exceptions are handled consistently.

3. Active Learning from Resolved Case Outcomes

Flow: OpenText Core Case ? Prodigy

Closed or resolved cases in OpenText Core Case can be exported to Prodigy as labeled or partially labeled training examples. Final case outcomes, resolution codes, and investigator decisions become high-value ground truth for retraining models that support classification, prioritization, or next-best-action recommendations.

Business value: Continuously improves model accuracy using real business outcomes and reduces manual labeling effort by reusing historical case decisions.

4. Document and Image Annotation for Case Intake Automation

Flow: OpenText Core Case ? Prodigy ? OpenText Core Case

Organizations can use OpenText Core Case to collect incoming case materials such as scanned forms, photos, screenshots, or correspondence, then send samples to Prodigy for annotation. The resulting labels can train models that automatically classify intake documents, extract key fields, or detect document types before they are attached back to the case.

Business value: Reduces manual intake work, improves first-pass routing accuracy, and shortens case creation and triage times.

5. Compliance and Policy Violation Detection Workflow

Flow: OpenText Core Case ? Prodigy ? OpenText Core Case

Compliance teams can use OpenText Core Case to manage investigations involving policy breaches, audit findings, or regulatory reviews. Relevant emails, reports, and evidence can be labeled in Prodigy to train models that detect risky language, missing disclosures, or non-compliant content. When the model identifies a potential violation, OpenText Core Case can open or update a compliance case for review.

Business value: Improves early detection of compliance issues, supports audit readiness, and standardizes investigation workflows.

6. Customer Support Case Categorization and Prioritization

Flow: OpenText Core Case ? Prodigy ? OpenText Core Case

Support organizations can use historical customer cases from OpenText Core Case to build training datasets in Prodigy for intent classification, urgency detection, and issue categorization. Once deployed, the model can automatically suggest case categories, priority levels, or escalation paths back in OpenText Core Case.

Business value: Improves routing speed, reduces backlog, and helps support teams focus on high-impact cases first.

7. Feedback Loop for Continuous Model Improvement

Flow: Bi-directional

OpenText Core Case can provide ongoing operational data such as case status changes, investigator corrections, and final resolutions to Prodigy for periodic retraining. In return, updated model outputs from Prodigy can be used by OpenText Core Case to improve case handling rules, decision support, and automation triggers. This creates a closed-loop process where operational outcomes continuously refine model performance.

Business value: Keeps AI models aligned with changing business rules, improves long-term accuracy, and supports scalable process automation.

8. Expert Labeling for Complex or High-Risk Cases

Flow: OpenText Core Case ? Prodigy

For high-risk or complex cases such as legal disputes, fraud investigations, or sensitive customer escalations, OpenText Core Case can route selected case records to Prodigy for expert annotation. Domain specialists can label nuanced text, images, or supporting evidence to create specialized datasets that reflect real-world edge cases.

Business value: Captures expert knowledge, improves model performance on difficult scenarios, and supports more reliable automation in sensitive workflows.