Estimate Your Ability with IRT

Professional certification exams don't grade on raw percent-correct — they use Item Response Theory to infer a candidate's hidden ability θ from which specific items they got right or wrong, weighted by each item's difficulty and discrimination. This simulator draws random exam items with real 2PL parameters, answers each one with a genuine Bernoulli draw from the 2PL probability curve, and refits a maximum-likelihood ability estimate by gradient ascent on the log-likelihood after every answer — watch the estimate converge toward the candidate's true ability as the exam progresses, and see whether it clears the certification pass threshold.