Insurance support is still heavily manual.
- High call volumes and longer waiting times
- Repetitive policy and claim requests
- Inconsistent answers across channels
- Higher operational effort and cost
- Poor customer experience during peak hours
Policy & Claims Automation using Google CX Agent Studio. Secure conversations, faster service, and a customer journey that moves from first word to confident resolution.
The capstone starts with a familiar contact-centre problem and answers it with secure multi-agent orchestration.
Built for anytime service in one conversation; production availability depends on deployment.
Registered-detail changes become Pending offline-review records; there is no promised call, message, approval, or automatic application.
Demo claims, cases, renewals and requests are written to BigQuery-backed systems.
A front door for authentication and intent detection, a focused specialist layer, and a grounded backend for every response.
5 specialist functions · 8 routed agents
Verified access • Grounded responses • Modular design • Scalable customer journeys
Phone number, DOB, returned security question and correct security answer · no OTP
Policy details, benefits information and policy renewal
Claim status tracking and claim initiation
A genuinely new phone is pre-authentication; an existing phone requires a bound auth token and matching identity
Create and track a Pending offline-review record; no call, message, approval, or automatic application is promised
Internal case-writing agent; the outcome is a persisted case with conversation context
List, resolve and select the correct policy
User-triggered renewal support and guided next steps
Fallbacks, retries, validation and error handling
Primary journeys: policy services • claims • onboarding • update request • support-case persistence
Prompt guards inspect the caller’s request, callbacks enforce security around every model and tool call, and fallbacks recover safely when conversations or integrations fail.
before_model · Guards prompt injection, third-party claims and support-case intent.before_tool · Protected policy and claim tools require the authenticated phone, customer, allowed policies and signed token; update tools require the bound phone, customer and token. Trusted values are injected into protected calls.after_tool · Saves only authoritative successful results such as claim IDs, request IDs and persisted case IDs.after_model · Masks sensitive phone output and suppresses farewell or end_session while an accepted escalation has not been persisted.createCase is successful only when success:true and persisted:true; a generated case ID without persistence is rejected.CALLBACK_REGISTRY.requestUpdate before a masked readback and explicit confirmation.Expected deterministic flow
Expected deterministic flow
15 customers across Motor, Health and Life—with active and lapsed policies and mixed claim states.
Mocked tools were migrated to a BigQuery-backed Cloud Run function, then verified for behavioural parity before real write operations were added.
The presentation supplies these values as placeholders for final evaluation. They are visualised as targets—not measured production results.
Illustrative target for correct intent routing
Illustrative target for successful verification
Illustrative target for journeys without human help
Illustrative target for end-to-end evaluations
Illustrative share resulting in a support case
Illustrative numeric benchmark, not a percentage
All six scorecard figures are placeholders reproduced from the deck; they must be replaced after the planned evaluation run.
Immediate response for routine requests
Grounded answers from APIs and data
More journeys automated
Context preserved across the journey
Anyone can find a phone number or birthday. The assistant therefore raises a request instead of changing registered details during a call.
Conversational AI Voice Agent for Insurance
Secure conversations. Faster service. Better customer experiences.
Final Project Presentation | EXL Internship 2026
Choose a scenario. The request is loaded into the Happy Insurance assistant—review it, then press Send. Text is the assured presentation path; audio remains available pending the final live microphone smoke test.
Use your own words. The assistant detects the intent and routes the journey.
Phone number, DOB, returned security question and correct security answer—no OTP. A genuinely new phone is pre-authentication; an existing phone requires a bound auth token and matching identity.
Claims, support cases and requests are registered with a reference ID and stored status.
Choose any scenario above to prepare the assistant.
Nothing is auto-sent—you remain in control.