// scoring · evidence before claim

How a score is computed.

OpsTicket returns a work-sample score, not a quiz result. You work a real ticket in a command environment; the server computes a weighted rubric score from the evidence your commands and their output leave behind. This page states the basis for that score, and the limits it does not exceed.

the rubric · deterministic weighted rubric

Five weighted dimensions.

Each scenario has an expert solution path and five bounded rubric dimensions. The weights below are fixed and sum to 100. The same weights are applied on every attempt, so two candidates who do the same work get the same score.

Accuracyweight 30

Methodologyweight 25

Tool fluencyweight 20

Communicationweight 15

Real-world fitweight 10

Contract version 2026-08-02.3 · rubric v2.0

who computes the number

The server does the arithmetic.

Evidence, not opinionThe inputs are terminal commands and terminal output. An automated evaluator reads that evidence and returns bounded bands; it does not hand back the final grade.

Server-side aggregationThe server multiplies the bounded evidence by the fixed weights and produces the score. The number is recomputed from evidence, so a client cannot submit its own score.

Bounded bands on the terminal pathOn the terminal grading path the evaluator is held to 0 to 3 bands per dimension, which the server weights to a 0 to 100 total.

what this is not

The limits, stated plainly.

The assessment terminal is a client-side simulated command environment, not a container, virtual machine, or PTY. Evidence is command-and-output entries, not per keystroke.

There is no webcam monitoring or attention proctoring.

A verified certificate is a server-issued record tied to a SHA-256 evidence hash. No external chain, public registry, or separate signing system is claimed.

The score is a hiring signal from a bounded rubric. It is not a guarantee of on-the-job performance and it does not replace your interview.

Verify a certificate →Try a demo scenario More questions