AI Data Analysis & Decision Support

Business questions are rarely posed as SQL. The aim of Insight Engine is to let the person asking get a checkable answer — not just a number, but how the number was derived.

Pulling data waits on the data team’s queue, and by the time an answer arrives the question has often gone stale. Yet auto-generated analysis that cannot state its definitions and sources is not usable for a decision.

• 01

  Natural-language querying

  Translates a business question into a query and shows the query it generated.

• 02

  Consistent definitions

  Metric definitions live in one place, so the same name means the same thing.

• 03

  Anomaly detection

  Surfaces deviations proactively and suggests possible contributing factors.

• 04

  Trend reading

  Describes change over time and states the uncertainty around it.

• 05

  Report generation

  Produces structured periodic reports carrying their data provenance.

1. 01

   Operating review

   Answers questions on sales, inventory and cost quickly.

2. 02

   Anomaly monitoring

   Detects metric deviation and flags likely causes.

3. 03

   Periodic reporting

   Drafts weekly and monthly reports automatically.

• Data warehouse
• Operational databases
• BI and reporting tools
• ERP and order systems
• Messaging channels

Access is read-only by default; writes require separate authorisation. Row- and column-level permissions follow role, and sensitive fields can be masked. The generated query is visible to the user, so results are checked rather than trusted blindly. Private deployment is supported.

In private preview for invited customers. Metric definitions must be settled first — without that, results cannot be relied on.

• 03Enterprise Knowledge & Internal Q&AKnowledge FabricTurning documents, policies, project material and system data into enterprise capability that can be searched, cited and kept current.Enterprise Delivery
• 04AI Workflow AutomationFlow EngineOrchestrating triggers, models, knowledge, tools, approvals and business systems into a process that keeps running.Private Preview
• 02Enterprise AI Agent StudioMoShun Agent StudioDesigning AI agents around real roles, processes, knowledge and system interfaces — agents that step into the flow of work rather than sit beside it.Enterprise Delivery

• 2026-06-28Research DirectionRetrieval: treating provenance as a first-class concernInside an organisation an answer without provenance cannot be trusted. We put citation ahead of raw retrieval quality.
• 2026-03-16Private PreviewInsight Engine: analysis is meaningless before definitions settleThe most common blocker in preview is not model capability but the same metric meaning different things in different departments.