← All Case Studies
Case Study

Mystica — Test Automation & QA

Quality Engineering · React Native · Node.js

I helped the team ship faster, with confidence. Mystica is a mobile app for live, paid chat sessions, where a single glitch can drop a session or mischarge a customer. As part of the engineering team, I drove its automated quality — cross-platform end-to-end coverage of the core money-making flows, load testing to find the limits before customers do, and the test strategy behind it — so changes go to production protected, not on a hope.

1
cross-platform E2E suite I co-built, iOS & Android
100
concurrent users in load testing
~40%
redundant test cases cut by smart design
E2E Automation

Stood up the cross-platform E2E framework — including Android support and the device runners — and authored many of the core-journey flows, passing on iOS and Android

Load Testing

Designed and ran load & stress testing up to 100 concurrent users to find the system’s real capacity limits

Test Strategy

Created the test-design model that focuses effort on the highest-risk flows and cuts redundant cases

Release Gate

Wired the suite into an automated gate that runs on every change — so nothing reaches production without passing the core-flow checks

The Challenge

  • Live chat and payments with timing-sensitive edge cases — easy to break, hard to catch
  • No end-to-end safety net for the core flow customers pay for: request → chat → bill → end
  • No visibility into how the system would hold up under real load

What I Did

  • Built out the cross-platform end-to-end suite covering the real journeys — sign-in, chat, payments, onboarding, and edge cases — passing on iOS and Android
  • Wired it to run automatically on every release, so regressions are caught before they ship
  • Designed and ran load & stress testing to find capacity limits before customers hit them
  • Set the test strategy — focusing effort on the highest-risk flows and cutting redundant cases

The Impact

  • Confidence to ship every release — the team trusts the core money-making flows are protected, so changes go out without second-guessing
  • Faster dev cycles — automated checks replace slow manual regression passes, so fixes and features ship sooner
  • Bugs are caught before release, not after — less time firefighting in production
  • Capacity limits surfaced early, so scaling decisions are made on data, not surprises

Smart Test Design

Rather than test everything blindly, I built a test-design model that cut redundant cases by ~40% — while keeping exhaustive coverage on the highest-risk chat & billing flows, where a single missed combination can drop a paid session. Less busywork, more protection where it counts.

Maestro Locust React Native / Expo TypeScript Jest GitHub Actions CI Self-hosted iOS runner PubNub Stripe

Want something built or hardened to this standard?

Get in Touch