A timed mock matching one recruiter's real, documented test format.
TCS, Infosys, Wipro, Cognizant, Zoho, Google, and Microsoft don't publish real past papers - every 'previous year questions' site online is a crowd-sourced reconstruction, not a leaked exam. So instead of scraping unreliable third-party content, every question here is original and hand-written, built to match what's actually documented about each company's section order, topic weighting, and timing. Live for all five mass-recruiter tracks: TCS, Infosys, Wipro, Cognizant, and Zoho.
TCS NQT format
English, Quantitative Aptitude, and Programming Concepts, back to back under one continuous clock - then a coding round.
Infosys format
Mathematical Ability, Logical Reasoning, Verbal Ability, Pseudocode Ability, and Puzzle Solving - 54 questions in one continuous 100-minute test.
Wipro NLTH format
One continuous 20-question Aptitude section, then a real, timed 200-300 word Essay Writing round - AI-graded on structure, clarity, and relevance to the assigned topic.
Cognizant GenC format
Numerical, Logical Reasoning, and Verbal Ability - 80 questions in one continuous 100-minute test - then a Technical Assessment (coding + SQL) and a real spoken Communication Assessment.
Zoho format - no multiple choice
Zoho's real written test is paper-and-pen with no options at all, so this is the one company here with typed short-answer questions instead of MCQs - Aptitude and C Programming (predict the exact output), each graded against a real answer, not options - then two coding rounds.
Spoken Communication Assessment
Reading Sentences, Repeat Sentences, and Jumbled Sentences are scored by real speech-to-text compared word-for-word against the target text; Question & Answer and Storytelling are AI-graded on relevance and coherence, since those have no fixed answer.
Sequential timed sections
One section's timer runs out (or you submit early) and the mock auto-advances to the next, matching how the real test actually runs.
Programming Concepts, Pseudocode & C output-prediction
New theory and code-tracing pools built just for this module - OOP, data structures, complexity, SDLC, step-by-step pseudocode traces, and real C snippets, every one independently re-executed and verified (the C pool compiled and run with real gcc).
Deep-links, not duplicate engines
Coding and SQL rounds hand off straight into Coding Challenges and SQL Challenges, instead of a second, duplicate code-execution engine.
Every MCQ and short-answer section is zero-AI and fully client-scored, same discipline as Aptitude & Reasoning - every question and explanation is hand-authored and verified, never AI-generated or scraped (Zoho's short-answer questions are graded by exact/numeric match, not multiple choice, since that's what its real test actually uses). Two rounds are the exception: Wipro's Essay Writing (an open essay has no single reference answer to check) and Cognizant's spoken Question & Answer / Storytelling items (no fixed answer either) are graded by AI on structure, clarity, and relevance - word count and transcript-to-target accuracy are always computed exactly, never AI-judged.