Helport AI Copilot
AI Systems | Human–AI Interaction | Product Design
Skill
AI Product Design, UX Research, System Architecture, Prompt Design, Prototyping, Evaluation Design
Tool
Figma, LLM APIs, Prompt Engineering Frameworks, User Testing
Year
2025
Helport AI Copilot is a human-centered intelligent assistant designed to support users through complex, high-cognitive workflows. Rather than functioning as a generic chatbot, it operates as a structured copilot that clarifies context, breaks down tasks into manageable steps, and provides transparent, actionable guidance. The project focused on designing AI as a collaborative partner—balancing automation with user control to build trust and reduce abandonment during difficult processes. I led the interaction design and system logic, defining how the AI interprets inputs, structures responses, and adapts to varying levels of task complexity. Through iterative prototyping and testing, the system was refined to prioritize clarity, progressive guidance, and decision support over raw information output.
Market Research
Problem Statement
Member Confusion
Students and members struggle to understand their coverage, face inconsistent answers, and often experience delays or denials that impact their health.
Agent Overload
Agents are overwhelmed by fragmented systems and complex policy variations, leading to long calls, frequent escalations, and compliance risks.
Systemic Inefficiency
Health insurance support remains slow, error-prone, and costly—driven by outdated tools, repeated calls, and preventable regulatory violations.
Solution Proposed
-
Empowers agents with instant, plan-specific answers triggered by live call transcription.
-
Uses guided speech navigation to walk agents through complex, multi-step processes.
-
Detects and flags HIPAA/CMS risks during live calls, providing real-time corrective prompts to avoid legal violations and protect star ratings.
-
Generates post-call summaries outlining key issues, coverage info, and next steps
