Skip to content
December Code

What is behind the practice

Repository bank inventory checked 16 Nov 2026. The deployed active bank can differ while a correction is being published. 3,000 questions across 20 subjects. No daily cap on a paid plan does not promise fresh questions forever.

Score calibration: not yet established

Scores describe sampled MCQ answers, not hiring probabilities. Evidence labels reflect counts and topic coverage. We do not currently publish hired-cohort benchmarks or placement predictions; those require consented outcome data and independent validation. Self-marked answers do not influence topic recommendations.

Published source bank depth
SubjectQuestionsMCQsTopicsTopics under 3 questions
DSA350326140
SQL400385120
Aptitude300300170
OOP140125150
DBMS140124140
Python45044490
Java45044390
JavaScript484280
C484280
C++484280
C#484280
Go484280
Rust484280
Kotlin484280
PHP484280
Ruby484280
Swift484280
Operating Systems1009480
Computer Networks1009480
Verbal Ability908680

Content review

The latest expansion adds 1623 original MCQs across 10 subjects, with more depth in Python, Java, SQL, DSA and aptitude. We ran 1,149 code questions and checked 183 quantitative answer derivations using Python 3.14, Java 17, PostgreSQL 17 and 18. Choices and similarity signals received AI review. Independent human moderation remains a separate step; these checks do not establish a hiring benchmark.

The current QC run flags 5409 review warnings and 126 error-worded keys for human inspection. These are signals, not confirmed errors. We review similarity, distractor plausibility, answer-format cues and explanations before correcting or retiring a question.

5533 current signals have a recorded Codex source review. These AI reviews are tied to the exact source revision and reopen when it changes. They include preservation checks on unchanged historical questions. They do not certify every answer by execution or replace independent human moderation.

Correction and review log

12 November 2026 · SQL-005 and SQL-034

Corrected the distinction between byte length and character count in MySQL. LENGTH counts bytes; CHAR_LENGTH counts characters. The wording and explanations now match the MySQL reference. Existing answer keys and option text were preserved.

11 November 2026 · KT-004

Clarified that a prior null check can permit a smart cast. The answer key was retained after checking the exact listing with Kotlin 2.3.10.

11 November 2026 · PHP-045

Balanced answer formatting so the correct option is no longer the only option detected as code. The stored answer text was preserved, keeping existing attempts comparable.

Report a question or explanation →