Conversation Design for AI-driven Benefits Verification
Built a visual conversation builder for Infinitus's AI-driven insurance verification system, enabling faster automation of complex workflows through reusable dialogue components.
Background
Infinitus faced a critical challenge while scaling their benefits verification automation.
While their AI system could handle basic calls, complex workflows required significant human operator intervention.
In order to build and debug these complex conversations at scale, conversation designers needed to make and manage flows.
The existing tool was becoming a bottleneck — conversation designers struggled with an unintuitive interface, lacked visual feedback, and could not easily reuse components across different flows and customers.
Key challenges
- add_circleConversation designers struggled with unintuitive interface
- add_circleLacked visual feedback
- add_circleCould not easily reuse components across flows and customers
- add_circleComplex workflows required significant human operator intervention
Outcomes
- add_circleLed design and engineering to overhaul the conversation builder tool, enabling users to convert new SOPs into automated dialogues >30% faster
- add_circleEnabled conversation designers to flatten conversations, opening the door for more sophisticated automation
- add_circleImproved response time to client requests by >20%
- add_circleStandardized component implementation, including providing automatic usage metrics for product and engineering
- add_circleEnabled rapid iteration on conversation designs through improved visual feedback and reduced QA time by >20%
Business Impact
Improved Efficiency
- add_circleReduced time to create new conversation flows
- add_circleDecreased maintenance overhead for existing flows
- add_circleEnabled rapid testing of automation strategies
Better Scalability
- add_circleStandardized components reduced implementation time
- add_circleReusable patterns accelerated new development
- add_circleImproved visibility into system behavior
Enhanced Quality
- add_circleReal-time preview caught issues earlier
- add_circleStandardized validation improved data quality
- add_circleBetter visualization reduced implementation errors
I am a demo title
I am a demo description (maybe I belong after the demo?)
I am a demo description (maybe I belong before the demo?)
Solution
Redesigned the conversation builder from the ground up, focusing on three key innovations
Visual Flow Editor
Improved visualization and editing of conversation flows
- add_circleCreated intuitive visual interface for designing complex insurance verification dialogues
- add_circleVisually map and modify conversation flows
- add_circlePreview utterances and responses in real-time
- add_circleReuse common dialogue patterns across different insurance providers
- add_circleValidate outputs against known data formats
Standardized Component Library
Increased consistency and reusability of dialogue components
- add_circleDeveloped comprehensive set of reusable dialogue components
- add_circleEnforced consistent implementation patterns
- add_circleProvided automatic usage metrics
- add_circleSimplified complex interaction patterns into reusable blocks
- add_circleEnabled rapid prototyping of new conversation flows
Output Management System
Improved data quality and system integration
- add_circleBuilt flexible system for managing conversation outputs
- add_circleEnforced data validation rules at design time
- add_circleProvided clear visibility into data dependencies
- add_circleEnabled cross-organization output standardization
- add_circleSimplified integration with downstream systems
I am a demo title
I am a demo description (maybe I belong after the demo?)
I am a demo description (maybe I belong before the demo?)
Process
User Research
- add_circleConducted extensive shadowing sessions with operators and conversation designers
- add_circleInterviewed >20 users across different roles to understand pain points
- add_circleCreated detailed workflow maps to identify optimization opportunities
Iterative Design
- add_circleDeveloped multiple design iterations with stakeholder feedback
- add_circleBuilt high-fidelity prototypes for user testing
- add_circleConducted usability studies to validate design decisions
- add_circleIterated based on quantitative usage data
Implementation Strategy
- add_circleCreated phased rollout plan to minimize disruption
- add_circleBuilt adapter layer to support gradual migration
- add_circleDeveloped comprehensive documentation and training materials
- add_circleEstablished new design patterns for future automation tools