Resources / User Guide / Core Products
Agentic Analytics
Use Agentic Analytics to run governed agents on validated context, review recommendations, inspect evidence, and route approved action.
Operations leaders, reliability teams, data teams, and governance owners reviewing AI-assisted recommendations.
Before you start
What to have ready
Validated DataOps context or an approved data product.
A defined workflow such as anomaly review, planning, diagnostics, compliance, inventory, or knowledge support.
Clear reviewer and approver roles.
A baseline expectation for what should remain advisory versus executable.
Usage workflow
How to use Agentic Analytics
Select the context
Choose the asset, source, event, run, or data product that the agent should use as evidence.
Run an agent or activation pack
Start the agent workflow, review the generated plan, and confirm that inputs and assumptions are visible.
Review recommendation evidence
Inspect traces, source facts, confidence, rejected options, and operational impact before accepting a recommendation.
Use approval and audit surfaces
Keep recommendations advisory until a reviewer approves the action and the audit trail captures the decision.
Measure outcomes
Track accepted actions, feedback, run quality, and business results so the agent workflow improves over time.
Product areas
Where to work in TwinEdge
Agents
Agent registry and overview.
Agent Runs
Run history and status.
Agent Audit
Review records and decision traces.
Agent Evaluation
Evaluation sets and quality review.
DataOps AI
AI context and DataOps assist surfaces.
DataOps Agent Runs
Runs tied to DataOps workflows.
App links require an approved TwinEdge account and the right module access.
Related guides
Need help?
Get implementation guidance
Request an activation discussion when you want to prove one agent workflow against one source or asset class.
Talk to TwinEdge