Interview engineers inside a live incident.

Drop candidates into a breaking system. They ask for logs, metrics, and deploy history to hunt down the root cause, and you see exactly how they got there.

Try it now

Watch how they think, not what they memorized.

Whiteboard rounds reward rehearsal; there's an entire prep industry for them. An incident can't be rehearsed. What they ask for, what they ignore, and how fast they reach root cause: signal no amount of grinding can fake.

How it works
Three asks, each spinning the hologram: architecture, regional gateway logs, recent deploys. Root cause found in 11 minutes 42 seconds. The path itself is the score.
THE PROOF

The score is the path.

Every question asked, every artifact pulled, and every dead end, timestamped from page to root cause to fix. Measured, not judged. Read the whole investigation in three minutes.

See the scoring
YOUR STACK

Test on the systems you run.

Import the job description and the incident happens on your architecture: your services, your failure modes. Generalist by default, specialized when you want it.

See pricing
KafkaRedisPostgreSQLKubernetesRabbitMQElasticsearchDynamoDBgRPCNginxMemcachedCassandraS3+ your JD

A new interview for an
era of rehearsed answers.

Screen for real engineering judgment without spending an engineer-hour.

Want to try it yourself?

Free practice incidents for candidates at launch. No spam, one email when it opens.