Tuning a reward economy with real player data
Launch numbers are a guess. What matters is being able to change them without a release, and knowing which number to move.
Blog
Notes from the build.
What we learned shipping our own products and clients’ — engineering decisions, testing, and how we work. Plus the archive from our consumer apps.
The monitoring you set up before launch, not after
After an incident you will wish you had three specific things. All of them take an afternoon to add beforehand and are painful to add during.
CI that gates releases instead of just running tests
A pipeline that reports failures nobody blocks on is theatre. What we make mandatory, and what we deliberately let through.
What our AWS bill taught us about architecture
The invoice is a design document nobody meant to write. Reading ours line by line changed decisions we thought were settled.
Migrating a live database without downtime
The pattern is always the same: expand, migrate, contract. Where teams get hurt is doing all three in one deploy.
Push notifications people do not turn off
A permission prompt is a one-shot resource. What we learned about when to ask, what to send, and how quickly goodwill runs out.
Handling money in code: never use a float
0.1 + 0.2 does not equal 0.3, and your payout ledger will eventually notice. How we represent money and why rounding is a product decision.
Offline-first on mobile: what actually breaks
Everyone agrees apps should work on a bad connection. Far fewer have decided what happens when two devices disagree about the truth.
What we check before adding a dependency
Every package is code you did not write, will not read, and now maintain. Five questions we ask before typing install.