Readiness score
Preparation Readiness is a score from 0 to 100, shown with a band from 0 to 5. It describes how much evidence you have produced across twelve interview competencies. It is not a probability of passing and it is never presented as one.
The twelve competencies
Requirements, scoping, estimation, data modeling, API design, architecture, scalability, reliability, security, trade-offs, follow-ups and communication. Each has its own score, and the overall score is a weighted combination of the competencies that have been tested. A competency with no evidence is shown as not tested rather than as zero.
The evidence ladder
Every completed task writes an evidence record that says how you practiced. Kinds of evidence are ranked from weakest to strongest:
| Evidence | Examples | Highest band it can support |
|---|---|---|
| Self-reported | Your diagnostic confidence ratings | 2 |
| Reading | Studying a question walkthrough | 2 |
| Flash cards | A spaced recall deck | 3 |
| Quiz | A scored quiz or the diagnostic items | 3 |
| Recall | Rebuilding a design from memory | 4 |
| Timed design | A design done against the clock | 4 |
| Evaluator | An Evaluator AI run on your design | 5 |
| Mock interview | A completed AI mock with a debrief | 5 |
| Real interview | A debrief with a known outcome | 5 |
Stronger evidence counts for more, and each kind has a cap on how far it can lift a competency on its own.
Ceiling bands
The band you can display is capped by the strongest kind of evidence you have. If everything so far is reading, the score cannot show more than band 2, no matter how much you read. Reaching band 5 requires evaluator feedback, a mock interview or a real interview.
Evaluator AI counts on its own. While a sprint is active, every design you submit to Evaluator AI records evaluator evidence from the rubric scores it returns, whether or not the run was on your plan, and an Evaluator AI task planned for that question is marked done.
The bands are labelled:
| Band | Label |
|---|---|
| 0 | No evidence yet |
| 1 | Starting |
| 2 | Building |
| 3 | Developing |
| 4 | Strong |
| 5 | Interview ready |
When a ceiling is holding your band down, the Readiness screen names the next unlock: the single kind of practice that would lift it, such as one timed design. It names an act, not a number of points.
Decay
Evidence fades. Each kind has a half-life, from about a week for reading to two months for a real interview, so a skill you practiced once three weeks ago counts for less than one you practiced yesterday. A competency with no fresh evidence for two weeks is marked stale.
Confidence
Alongside the score, readiness shows a confidence of low, medium or high. It rises with the amount of evidence, the variety of evidence kinds and how recent it is. A high score with low confidence means the plan does not yet know you well.
Per interview
Each interview has its own readiness, weighted toward the competencies that interview is likely to test. The overall figure on Mission Control covers the whole sprint.
Reproducibility
Every snapshot stores the evidence it was computed from and the version of the formula used, so a past score can always be explained exactly. When the formula changes, it ships as a new version and older snapshots keep their original meaning.