Contactless Guest Check-in Mini App

I. System Context & Constraints

Technical reality: An operator runs an ecosystem that automates unmanned accommodation — vacation rentals and hotels. Guests checked in through an in-facility tablet or a one-time web flow, leaving no persistent channel: if a guest needed their door PIN or Wi-Fi password mid-stay, they had to re-navigate the web check-in from scratch, and the operator had no way to re-engage them after checkout.

Director’s constraints: The target market lives inside one dominant messaging app, so the experience had to be native to it and fast on mobile networks. Guest data (key codes, Wi-Fi credentials) demanded careful PII handling, and the architecture had to stay loosely coupled so the same product could later extend to other messaging platforms for international guests.

II. Critical Path & Architectural Choices

We built the experience as a messaging-platform Mini App (LIFF) with a backend-for-frontend (BFF) and AWS infrastructure defined in CDK. Initializing LIFF captures a stable user ID on every session, which converts a one-time check-in into a re-engageable, persistent guest relationship — the core strategic goal.

The flow is deliberately shallow: a booking-code lookup returns the reservation, and detail screens surface Wi-Fi, key/entry, and instructions with one-tap clipboard copy, redirecting to the existing web check-in only where registration is required. PII transfers are encrypted with AWS S3 KMS, and the frontend/backend are kept loosely coupled so a future WhatsApp or WeChat deployment reuses the same BFF.

III. Governance, Risk & Team Topology

This was a multi-party delivery — a bridge consultancy handling PM, UI/UX, and QA alongside our engineering team. I governed the milestone plan across that boundary: setup, the reservation-and-display build, a staged client review, detail screens and redirect, full UAT, and production deployment with handover.

Risk was managed through staged reviews on a staging environment, structured feedback collection and re-test loops, iOS and Android UAT before any go-live decision, and a production smoke-and-performance check followed by a completion report and documentation handover.

IV. Quantifiable Outcomes