QA Engineer (Crypto Casino / AI-Native Team)

About Us
We're a small, fast-moving team building a crypto casino — real-money games, instant crypto deposits and withdrawals, and the infrastructure behind them. We run an AI-native engineering organization: much of our code is written, reviewed, and released by AI agents orchestrated through an automated pipeline. Releases are cut continuously and deployed to dedicated QA environments before reaching production.
This changes what QA means here. The bottleneck isn't writing test cases for human developers — it's being the sharp pair of human eyes that catches what automated pipelines and AI reviewers miss. In a casino, a missed bug isn't a bad user experience — it's money paid out wrong.
The Role
You will own the quality gate between "the pipeline says it works" and "a real player's money moves correctly." Every release cut lands in a QA environment; your job is to break it before players do.
What you'll do day to day
Test release candidates end to end. Work through QA environments on every release cut: play the games, exercise the product flows manually, verify new features against their specs, and hunt for regressions in existing behavior.
Validate money flows deeply. Crypto deposits and withdrawals, bet placement and settlement, win/loss accounting, bonuses and promotions, balance updates across game rounds. You need to understand what should happen — both on-chain and in our ledger — and notice when the two disagree, even by a little. Small discrepancies in a casino are never small.
Verify at the database level. A game can look right while the accounting underneath is wrong. You'll query our databases directly to confirm that what the player sees matches actual state — balances, bet records, payout history, bonus ledgers.
Think in flows, not screens. Trace a player journey across the frontend, the game engine, the wallet, the chain, and the database. Flag when a flow is technically functional but doesn't make sense — confusing states, exploitable edge cases, race conditions around concurrent bets, anything a motivated player could abuse.
Work with Claude Code as a daily tool. Use it to explore the codebase, understand what changed in a release, reproduce bugs, write one-off verification scripts, and file precise, well-scoped bug reports that our AI engineering pipeline can act on directly. The quality of your bug report determines the quality of the fix.
Close the loop. A bug you file goes into the same pipeline that ships features. You'll verify fixes, confirm root causes were actually addressed, and push back when they weren't.
What We're Looking For
Hands-on manual QA experience: exploratory testing, regression testing, writing clear reproduction steps.
Working SQL skills — comfortable querying a database to verify application state, not just reading dashboards.
Real understanding of crypto fundamentals: wallets, transactions, confirmations, and what can go wrong at each step. You've used crypto products as a real user.
Practical experience with Claude Code (or strong experience with AI coding agents and the willingness to go deep fast). You should be comfortable letting an agent navigate a codebase for you and directing it precisely.
Strong written English. Your bug reports are consumed by both humans and AI agents — ambiguity in, garbage out.
A genuinely adversarial mind. You think like a player trying to cheat the house: double-spends, disconnect-mid-bet, bonus abuse, timing exploits. You assume things are broken until proven otherwise, and you enjoy proving it.

Work arrangement
Yes