The AI Recruitment Glossary
Master the language of the future. The essential guide to the terms, tools, and technologies revolutionizing global talent acquisition.
Context Engine
MockWin's Context Engine is the pre-interview AI system that reads a candidate's resume, maps it to a job description, and builds a personalised adaptive question tree before the interview session begins. Unlike static question banks, the Context Engine ensures every MockWin interview is calibrated to that specific candidate and role.
Smart Resume Parser
MockWin's Smart Resume Parser is the NLP module that extracts structured skills, experience, tools, certifications, and project data from uploaded resumes to power personalised interview sessions. Unlike basic text extractors, it applies semantic weighting — distinguishing between a tool mentioned once in passing versus a skill listed across multiple roles.
JD Matcher
MockWin's JD Matcher compares a candidate's parsed resume profile against the active job description to produce a three-category alignment map: matched skills, partial matches, and gaps. This map drives MockWin's adaptive question generator — matched areas receive validation questions, partial matches receive transferability probes, and gaps receive exploratory questions.
Adaptive Questioning Logic
MockWin's Adaptive Questioning Logic is the real-time AI capability that evaluates each candidate response and dynamically selects the next question — escalating depth for strong answers, probing gaps for weak ones, and pivoting topics when knowledge limits are reached. Unlike fixed interview scripts, it makes every MockWin session unique to the candidate's demonstrated knowledge.
Drill-Down Architecture
MockWin's Drill-Down Architecture is the multi-level question branching system that enables up to three consecutive follow-up questions on a single topic. Level 1 tests familiarity, Level 2 tests applied knowledge, and Level 3 tests edge-case depth — separating practitioners from candidates who have memorised standard interview answers.
Contextual Memory
MockWin's Contextual Memory retains specific statements and named claims from earlier in the interview session and references them naturally in later questions — creating conversational continuity that mirrors a skilled human interviewer. It operates within a single session and does not persist between separate interviews.
Domain Classifier
MockWin's Domain Classifier identifies the primary and secondary professional domain of a candidate or role — such as engineering, product, or data science — to route the interview to the correct question bank and evaluation rubric. It supports dual-domain assignment for hybrid roles such as Technical Product Manager (Product + Engineering) or Revenue Operations (Sales + Data).
Seniority Calibration
MockWin's Seniority Calibration matches interview question difficulty, drill-down depth, and evaluation standards to the seniority level of the target role — ensuring junior, mid-level, and senior candidates are not assessed against the same bar. It operates on a five-band scale from Entry to Executive and supports mid-session dynamic recalibration via Adaptive Questioning Logic.
Nice-to-Have Keywords
Nice-to-Have Keywords are the preferred but non-essential skills and qualifications extracted from a job description by MockWin's JD Matcher — identified through language markers such as 'preferred,' 'a plus,' or 'nice to have.' Candidates whose resumes match these keywords receive additional exploratory interview questions, creating an automatic differentiation layer between candidates who meet the baseline and those who exceed it.