Question 1

What is Scenar.io?

Accepted Answer

Scenar.io is a practice site for DevOps and SRE debugging interviews. You get a broken Linux server and an AI interviewer. You say what you would check, the AI answers as the server, and you work toward the root cause, a short-term fix and a long-term prevention.

Question 2

Is it a real terminal?

Accepted Answer

No. Nothing runs on a real server. You describe in plain words what you would do, such as "check disk usage" or "look at the nginx logs", and an AI writes the output a real server would plausibly show, based on a hidden state defined for that scenario. It is a conversation, not a terminal emulator, so exact syntax does not matter.

Question 3

How does a round work?

Accepted Answer

Pick a scenario and a difficulty. The interviewer sets the scene, then you investigate by saying what you would run. The AI replies with simulated output and the app tracks the clues you find. A round moves through three stages: find the root cause, apply a short-term fix and propose a long-term prevention. There are also verbal rounds, where you answer conceptual questions in your own words and the AI scores them, and sandbox rounds, where you explore a simulated system with no fixed bug.

Question 4

Do I need an account?

Accepted Answer

No. You can start a round without one. Signing in with GitHub saves your progress to your account, and sandbox rounds need a sign-in. We do not ask GitHub for your email address.

Question 5

Is it free?

Accepted Answer

Yes. The free tier gives you 5 debugging rounds, 3 verbal rounds and 2 sandbox rounds a month, with no card needed. Paid passes are coming but are not on sale yet. They will be one-time purchases with no auto-renew and a 14-day refund.

Question 6

What topics are covered?

Accepted Answer

There are 21 debugging scenarios across Linux system administration, networking, containers, security, storage and databases. Three are adapted from real public postmortems: GitLab's 2017 database loss, Cloudflare's 2019 regex CPU outage and Cloudflare's November 2025 config-file outage. Verbal rounds cover Linux, Kubernetes, Terraform, SRE practice, networking and system design.

Question 7

Can the AI be wrong?

Accepted Answer

Yes. The interviewer, the simulated server and the scores are all AI, and the AI makes mistakes: an output can be inconsistent and a score can be off. It is practice feedback, not an assessment or a certification, and it promises no interview outcome. Check a real command against its documentation before you run it on a real system.

Question 8

What happens to what I type?

Accepted Answer

Your commands and answers are sent to an AI model provider to produce the reply, and they are stored with your attempt together with the replies. Do not type secrets or personal data. The Privacy Policy lists what is kept, who processes it and how to ask for deletion.

Question 9

Why did I hit a limit?

Accepted Answer

Several limits can apply. Free accounts have monthly round allowances that reset at the start of each month. A session has a cap on turns, and a round has a cap on hints. A daily AI-use budget keeps the site affordable for everyone and resets at 00:00 UTC. Daily practice is once a day, rate limits slow down very fast requests, and Hard difficulty needs a pass. The message on screen names the limit you hit and, for limits that reset on a schedule, when.

Question 10

How do I report a bug or security issue?

Accepted Answer

Write to support@scenar.site with what you did and what you saw, and leave out secrets. Use the same address for a security issue; the disclosure details are in /.well-known/security.txt.