Cassidy AI Applied Automation

Build accurate, controlled, and practical AI-powered workflows.

Duration: 2 Days

Many business processes become slow because essential information is scattered across documents, inboxes, and workplace systems. Cassidy AI brings this information together through Knowledge Bases, configurable Agents, no-code Workflows, connectors, and organization-wide instructions, enabling teams to automate repeatable work while retaining human control.

This course focuses on designing dependable solutions that can review documents, retrieve approved information, structure business data, route exceptions, and support consistent decisions. The instructor has over 30 years of industry experience and will deliver real, industry-demanded content rather than an academic treatment of the subject.

Learning Outcomes

Participants will be able to:

Identify processes suitable for Cassidy AI automation
Organize trusted information within a Knowledge Base
Configure specialized Agents with clear instructions and restrictions
Build multi-step Workflows using triggers, actions, and conditions
Develop practical insurance and underwriting support processes
Apply human review, testing, security, and governance controls

Prerequisites

Basic familiarity with Cassidy AI
Understanding of routine insurance and underwriting processes
Access to an approved Cassidy AI workspace
Approved non-confidential training documents
No programming experience required

Training Outline

Planning Cassidy AI Solutions
1. Process selection
  1. Repetitive administrative activities
  2. Document-intensive processes
  3. Information retrieval activities
  4. Review and approval processes
2. Process mapping
  1. Required inputs
  2. Processing stages
  3. Decision points
  4. Exceptions and referrals
  5. Expected outputs
3. Responsibility boundaries
  1. Automated activities
  2. AI-assisted activities
  3. Human-controlled decisions
  4. Restricted actions
  5. Process ownership
Building Trusted Knowledge and Agents
1. Knowledge Base organization
  1. Underwriting guidelines
  2. Risk appetite documents
  3. Policy wordings
  4. Authority limits
  5. Operating procedures
2. Knowledge quality
  1. Approved information sources
  2. Document versions
  3. Effective dates
  4. Conflicting information
  5. Outdated information
3. Agent configuration
  1. Agent purpose
  2. Required knowledge
  3. Operating instructions
  4. Output structure
  5. Escalation requirements
4. Agent controls
  1. Missing-information handling
  2. Assumption prevention
  3. Source-grounded responses
  4. Decision limitations
  5. Confidentiality restrictions
Designing Multi-Step Workflows
1. Workflow structure
  1. Workflow objectives
  2. Required inputs
  3. Processing actions
  4. Conditions and branches
  5. Completion requirements
2. Workflow triggers
  1. Manual initiation
  2. Form submissions
  3. File uploads
  4. Email events
  5. Scheduled events
3. Workflow actions
  1. Knowledge retrieval
  2. Document analysis
  3. Information extraction
  4. Agent execution
  5. Notifications and updates
4. Workflow controls
  1. Required-field checks
  2. Eligibility conditions
  3. Referral conditions
  4. Approval conditions
  5. Failure handling
Insurance and Underwriting Applications
1. Submission intake
  1. Document classification
  2. Applicant information extraction
  3. Coverage information extraction
  4. Exposure information extraction
  5. Completeness checks
2. Underwriting review
  1. Guideline retrieval
  2. Risk appetite alignment
  3. Risk information summaries
  4. Missing-information identification
  5. Exception identification
3. Referral preparation
  1. Referral reason
  2. Relevant guideline
  3. Supporting information
  4. Authority position
  5. Required decision
4. Renewal support
  1. Expiring policy information
  2. Exposure changes
  3. Claims developments
  4. Coverage changes
  5. Outstanding requirements
5. Business communications
  1. Submission acknowledgements
  2. Information requests
  3. Referral notifications
  4. Status updates
  5. Review summaries
Human Review and Approval
1. Review requirements
  1. Risk-based review
  2. Exception-based review
  3. Authority-based review
  4. Communication review
  5. Final approval
2. Review outcomes
  1. Approval
  2. Rejection
  3. Correction request
  4. Additional information request
  5. Manual processing
3. Approval records
  1. Reviewer identity
  2. Review outcome
  3. Supporting information
  4. Reviewer comments
  5. Completion status
Testing, Governance and Deployment
1. Agent testing
  1. Normal requests
  2. Incomplete information
  3. Conflicting information
  4. Restricted requests
  5. Escalation conditions
2. Workflow testing
  1. Standard processing paths
  2. Conditional paths
  3. Referral paths
  4. Approval paths
  5. Failure paths
3. Output validation
  1. Accuracy
  2. Completeness
  3. Consistency
  4. Source support
  5. Business-rule compliance
4. Governance controls
  1. Approved use cases
  2. User permissions
  3. Confidential information
  4. Human accountability
  5. Change management
5. Operational deployment
  1. Pilot implementation
  2. User responsibilities
  3. Performance monitoring
  4. Issue management
  5. Continuous improvement

Disclaimer

This course outline is provided as a general instructional guideline and does not constitute a fixed or contractually binding programme. The trainer reserves the right to amend, reorganize, substitute, expand, reduce, or omit topics when reasonably necessary to address participant requirements, organizational priorities, platform developments, operational constraints, or professional considerations, without prior notice.

Practical, connected learning

My wider training approach brings hands-on implementation and systems thinking together, connecting technology with real operational needs.

Cassidy AI Applied Automation