Artificial intelligence

Artificial intelligence in Quantemplate operates across four distinct areas:

Each is designed to enhance accuracy, efficiency, and control, while maintaining full transparency.

Trace

A silent safeguard that monitors workflow construction and prevents errors as users build pipelines. No shared data is required.

• Real-time warnings →
  Pre check data transformations during pipeline build, reducing the potential for errors.
• Permission alerts →
  Notify when other users – or robots – need access to resources such as reference data.
• Loop detection →
  Prevent infinite loops when chaining pipelines together.

Robot

The agentic layer of Quantemplate, automating routine workflows. This is a tightly focused set of tools, allowing whole processes to run autonomously and notifying the user of any exceptions. No shared data is required.

• Schedules →
  Request data on a schedule.
• Auto Run →
  Run a pipeline when data is received or an input updates.
• Auto Export →
  Export from a pipeline when a run completes; block and notify if validations fail.
• API Connections →
  Orchestrate connections to external systems.

Corrections

A controlled framework for improving data quality, with full visibility and auditability of all changes. Data is only shared explicitly with defined partners via the Chain of Custody.

• Auto Map Values →
  Use prior mappings as waypoints for future mapping suggestions.
• Validation corrections →
  Coming soon. Request corrections from partners and have those corrections applied automatically in the pipeline. Receive suggestions for corrected values based on decisions made within your organisation or across the network.

Network Intelligence

A recommendation layer that draws on network-wide patterns to suggest proven, industry-relevant workflows.

Safeguards include human-in-the-loop approval and a full audit trail, ensuring recommendations remain transparent, verifiable, and resilient to error or misuse.

Suggestions from activity within your organisation:

• Map Column Headers →
  Intelligent mapping suggestions based on column content, context, and mapping decisions made across your organisation. Coming soon: suggestions from mappings made across the network.
• Auto Map Values →
  Prior approved mappings within a pipeline act as waypoints for future mapping suggestions. Coming soon: suggestions from your whole organisation.
• Feeds →
  Naming conventions for submissions are suggested when another feed is connected.

Additional data network (coming soon):

• Joins →
  Logic-based join point suggestions for complex reconciliation work.
• Detect Headers →
  Automatic detection of data structures.
• Suggested workflows →
  Build pipelines faster with suggested configuration and end-to-end workflow recommendations.
• Network Intelligence Participation →
  All participation is controlled. Data sharing is opt-in via the Admin panel, allowing organisations to contribute to and benefit from collective intelligence.