Francis
An embedded financial AI experience with secure onboarding, streaming conversations and dependable account journeys across a production Expo app.
- Client
- Francis
- Year
- 2026
- Disciplines
- Mobile · AI · Backend
- Role
- We led core mobile product engineering and production hardening.
The problem
Francis needed to turn sensitive financial context into useful in-app guidance while preserving secure access, predictable navigation and a trustworthy experience across onboarding, subscriptions and account management.
The approach
We led delivery of the modular Tell Francis chatbot runtime with SSE streaming, persisted summaries and route-safe actions. We also engineered authentication, email verification, biometrics, onboarding, freemium and paywall gating, deep links, analytics and resilient account lifecycle flows using Expo Router, Zustand and TanStack Query.
The outcome
We hardened the production experience by preventing duplicate submissions, stabilising asynchronous mutations and navigation after account creation, and handling streamed-assistant failures clearly. Unit, integration and Maestro end-to-end coverage protected the journeys customers depend on.
Delivery highlights
- Led the modular Tell Francis chatbot runtime with SSE streaming and persisted summaries.
- Delivered secure onboarding, email verification, biometrics and account lifecycle journeys.
- Implemented freemium and paywall gating, deep links and account-event analytics.
- Hardened asynchronous mutations, duplicate-submission protection and route-safe actions.
- Covered customer-critical behaviour with unit, integration and Maestro end-to-end tests.
AMiSi Mobile Starter
