We think apps should fix their own bugs. TestFlight feedback comes in. An agent reproduces it, files the issue, writes the fix and runs it on a simulator before opening the PR. Our Cloud Simulators service is in preview. Here's the waitlist if you want to try it:

How can I achieve this on Android.... coming soon?

The fix was never the hard part. Reproducing a TestFlight crash on the exact device that threw it was. Nail that repro step and you've solved the real bottleneck for solo builders shipping on Expo.

That loop needs a trust boundary: the agent can propose a patch, but the PR should carry the original feedback, a deterministic repro, and the simulator trace that proves the fix. Otherwise “fixed” is just a successful code generation event.

Sick

this is too good for testing apps

reproducing the TestFlight bug before writing the fix is the part I’d trust most here.

Totally agree with this

does the simulator keep the exact TestFlight device state? otherwise the agent gets to fix a bug it accidentally made reproducible

The simulators look cool - if you want something that runs in WASM in the browser but still native and not DOM, this Open Source library may be interesting.

TestFlight geri bildiriminden simülatörde doğrulanmış PR'a giden akış, RN ekiplerinde QA turunu ciddi kısaltır. Cloud Simulators preview'ini deneyeceğiz. #React #ReactNative #Expo

Closing the loop from TestFlight feedback to reproduction, a fix, and a validated PR is a compelling agent workflow. The simulator verification step is especially important.

Running it on a simulator before the PR is the part that matters, everything before that is the agent grading its own homework. Caveat from my Expo app: two layout bugs only reproduced at XXXL text size in Italian. Default sim settings would have called both of them fixed.

the simulator recording is useful, but the reproduction step is the real receipt. does it rerun that exact path after fixing?