Prodigy - OpenText Core Digital Asset Management Integration and Automation

Common Integration Use Cases Between Prodigy and OpenText Core Digital Asset Management

1. Curated asset selection for model training

Data flow: OpenText Core Digital Asset Management to Prodigy

Marketing, product, and operations teams store approved images, videos, and documents in OpenText Core Digital Asset Management. An integration can automatically push selected assets into Prodigy for annotation when a machine learning project requires labeled training data. This is especially useful for computer vision use cases such as product recognition, defect detection, or content classification.

• Reduces manual export and file handling
• Ensures only approved, version-controlled assets are used for training
• Speeds up dataset creation for AI teams

2. Metadata-driven annotation queues

Data flow: OpenText Core Digital Asset Management to Prodigy

OpenText Core Digital Asset Management metadata such as campaign, product line, region, rights status, or content type can be used to create targeted annotation queues in Prodigy. For example, a retail organization can send only seasonal product images from a specific catalog segment to Prodigy for labeling, improving annotation relevance and reducing noise in the training set.

• Supports more precise labeling workflows
• Improves dataset consistency by using DAM metadata as filters
• Helps teams train models on business-specific content subsets

3. Human review of AI-labeled assets before DAM publication

Data flow: Prodigy to OpenText Core Digital Asset Management

After Prodigy is used to label assets or generate classification outputs, the reviewed labels and confidence scores can be sent back to OpenText Core Digital Asset Management as enriched metadata. This allows DAM users to search, govern, and publish assets based on AI-derived tags that have been validated by subject matter experts.

• Improves asset searchability and retrieval
• Creates a controlled path from AI output to enterprise content governance
• Supports downstream publishing and reuse of enriched assets

4. Closed-loop quality control for visual content

Data flow: Bi-directional

Organizations can use OpenText Core Digital Asset Management as the source of truth for product or brand assets, then send samples to Prodigy for defect detection, compliance labeling, or brand guideline checks. Once annotations are completed, results can be written back to the DAM to flag approved, rejected, or needs-review assets. This is valuable for manufacturing, retail, and regulated industries where visual quality must be checked before distribution.

• Creates a repeatable quality assurance workflow
• Helps identify non-compliant or low-quality assets early
• Supports faster approval cycles for large content libraries

5. Training data enrichment for content auto-tagging models

Data flow: OpenText Core Digital Asset Management to Prodigy, then Prodigy to MLOps or AI services connected to DAM

Enterprises often want to automate tagging of large asset libraries. OpenText Core Digital Asset Management can supply representative assets to Prodigy for manual labeling, and those labels can be used to train custom auto-tagging models. The resulting model can then generate suggested tags back in the DAM, reducing manual metadata entry for content operations teams.

• Reduces cataloging effort for large digital libraries
• Improves metadata consistency across teams
• Enables scalable AI-assisted asset management

6. Rights-aware dataset preparation for compliant AI training

Data flow: OpenText Core Digital Asset Management to Prodigy

OpenText Core Digital Asset Management often contains rights, usage restrictions, expiration dates, and regional permissions. An integration can ensure only assets cleared for machine learning use are sent to Prodigy. This is critical for organizations training models on branded content, customer-facing media, or third-party licensed materials.

• Prevents use of restricted assets in training datasets
• Supports auditability and compliance requirements
• Reduces legal and governance risk in AI programs

7. Feedback loop for content taxonomy improvement

Data flow: Prodigy to OpenText Core Digital Asset Management

As annotators in Prodigy identify recurring categories, edge cases, or misclassified content, those labels can be synchronized back to OpenText Core Digital Asset Management to refine taxonomy and metadata structures. This helps content teams improve search filters, folder structures, and tagging standards based on real-world annotation patterns.

• Aligns DAM taxonomy with actual content usage
• Improves discoverability across the enterprise
• Supports continuous metadata governance

8. AI project traceability and audit support

Data flow: Bi-directional

For regulated or high-stakes AI initiatives, OpenText Core Digital Asset Management can store the original asset, version history, and approval status, while Prodigy stores annotation decisions and model training labels. Linking the two systems provides end-to-end traceability from source asset to labeled dataset, supporting audits, reproducibility, and governance reviews.

• Provides a clear lineage between source content and training data
• Supports compliance, audit, and model governance processes
• Makes it easier to reproduce datasets and retrain models