Your AI worker for repeatable game QA.
Run repeatable gameplay checks, review the evidence, and build on what BugHunter learns about your game. We help with onboarding; your team can operate the product.
Your team runs it. We help you get started. We keep each studio's games and results confidential.
Every new build needs another round of checks
Every new build adds another round of checks. Give BugHunter a defined player flow and review the result with your team.
Your QA team can't cover everything before every release
Teams ship fast. QA headcount stays the same size. Something's always going to slip.
Build validation eats half your sprint
Running the same manual regression checks every build is slow, expensive, and still misses things. There's a better use of your team's time.
Bugs that reach players cost more than bugs caught in QA
1-star reviews, support tickets, Day 7 retention drops. A single bad update can take months to recover from.
Your next build can ship cleaner. We'll show you how.
How BugHunter works
It's simpler than you think.
Start with your game
During onboarding we help connect a supported build and device, any account or reset needs, and the expected behavior for the flows your team wants checked. After that, your team assigns the work.
BugHunter plays your game
BugHunter runs the agreed checks on a supported device, plays through the flow, and records the session so your team can see what happened.
Review findings across builds
Compare each agreed check against the expected behavior across builds, and review the findings and gameplay evidence with your team.
Your team reviews the findings
Video clips, device state, logs, and reproduction steps come together so your team can review each finding against the expected behavior.
How a BugHunter session works
An illustration of the steps in a single agreed run. Actual scope and timing depend on your game and the task.
Setup confirmed
Build, device, and scope agreed
Task running
BugHunter plays the agreed flow
Session recorded
Gameplay captured with logs
Evidence ready
Video, logs, and steps
Findings reviewed
Reviewed with your team
Illustration of a single agreed run. Scope and setup are confirmed with your team.
Grounded in your game's context
BugHunter works from your game's context: the instructions you share, the test data and expected behavior for each flow, and your QA team's feedback. Our team uses that context to adapt the worker to the checks you need.
Results depend on the game and the task. We agree the supported build and device up front, and your team reviews the findings and decides what needs attention.
Game instructions
How the flow is meant to work and what counts as correct.
Test data & expected behavior
The starting conditions and results your team agrees for each check.
QA feedback
Your team’s notes shape how the worker is adapted, run over run.
Curious what BugHunter finds in your game - specifically?
We'll agree a small scope on a supported build of your game and walk you through the findings.
No build upload needed to send this request.
What a first demo shows on your game
We agree a small scope, run it on a supported build, and review the findings together.
Regression checks that repeat
Run the agreed test cases against expected behavior and compare the results across builds.
Runs on supported devices
BugHunter plays through the agreed flow on a device and build we confirm with you for the task.
Reviewable findings
Each finding includes gameplay video and the assigned task so your team can decide what needs attention. Logs and reproduction detail depend on the run and setup.
A supported setup
We agree the build, device, and any account or reset needs for the demo before work begins.
Findings reach the team that acts on them
Keep findings connected to the build and the team that will act on them. We agree the review and issue handoff during setup; the planned MCP workflow brings requests and evidence into the developer's tools.
Grows with your builds
Keep the agreed checks running across upcoming builds as your game changes.
Real devices. Real purchase journeys.
Test in-game offers, store checkout, and reward delivery on real devices. Run purchase scenarios with platform sandbox accounts and test payment methods, capture what happened, and retest after the fix.
Release candidate
Repeat the tutorial, progression, and selected critical flows on the latest build.
Purchase and reward
Check checkout outcomes and the item or entitlement the game grants, including cancel, decline, and restart.
Hotfix retest
Replay a known issue and verify the developer's fix on the replacement build.
Sandbox purchase testing uses platform test accounts and test payment methods. This covers purchase flows, not all payment processing or production settlement. See the full example purchase check on How It Works.
When you don't know what to test - Scout does
Scout Mode explores a selected part of your game and captures mechanics, player flows, screenshots, and draft notes your team can review.
Use it to learn a new build before writing tests, then turn the observations into checks your team approves.
Your team runs it. We help you get started.
We help connect your game and prepare the first checks. Your team then chooses builds and tasks, reviews findings, and reruns checks. We keep each studio's games and results confidential.
Onboarding, with our help
We help connect a supported build and prepare the first checks and expected behavior.
Operated by your team
Your team assigns work and reviews findings with gameplay evidence, and decides what needs attention.
Confidential by default
Each studio's games, results, logos, and identities are not published.
See BugHunter work on your actual game
Send a game link and see BugHunter carry out a concrete check on your game, then see how your team runs it after onboarding.