Don't mark by hand, just review
Qdemy reads handwritten or typed answer scripts and turns them into scored, annotated, publish-ready results — to your rubric, in every subject.
Made for the way your institution works
Select one to see how Qdemy fits.
Qdemy for Universities
End-semester evaluation for 20,000 scripts without the six-week backlog. One rubric, one standard, every department — results published while the term is still fresh.
Qdemy for Exam boards
Consistent in Delhi, Dubai and Detroit. The same rubric applied at every centre, with anonymization and a complete audit trail for every mark.
Qdemy for Coaching institutes
Weekly test series marked overnight. Every student gets an annotated PDF with criteria-level feedback before the next class — not three weeks later.
Qdemy for Educators
Mark on screen with stamps, pen and highlights — or command the co-pilot in plain English and review its before/after diffs. Your expertise, minus the drudgery.
Qdemy for Students
Your own script back in days — annotated, with a score you can trace to the rubric and feedback that tells you exactly where the marks went.
Don't just get suggestions. Give commands.
Ask it anything in plain English — fix a model answer, balance a rubric, re-score with partial credit. It proposes a concrete before/after change you accept or discard. No prompt engineering, no context-switching.
Try it yourselfScore 300 papers in under an hour
The 5-stage pipeline reads every handwriting style, detects struck-through edits and rough work, aligns each answer to the rubric, applies scoring criteria, and emits an annotated PDF — automatically, for every paper.
Rubric designer
AI extracts sections, questions and Bloom's tags from the question paper. Marks reconcile live — catch mismatches before you freeze a version.
Every pen stroke, understood
From clean print to difficult cursive — 99% OCR accuracy, with strike-throughs and rough work detected automatically.
STEM-grade, natively
LaTeX in rubrics and model answers, diagram annotation against references, graph recognition with partial marks for correct structure.
Every student gets an annotated result
A fully annotated copy of their own script — scores stamped per question, feedback notes, and criteria-level justification. Not just a number.
Human expertise, digital precision
On-screen marking for human examiners — stamps, freehand pen, highlights and comments, all stored as structured data. Double marking with Chief Examiner moderation built in.
Your data never leaves your control
Enterprise-grade protection, anonymization and audit trails — compliant with DPDPA 2023.
Full anonymization
Evaluators — human or AI — see only ANON-XXXX IDs. Real identity is reconciled only at publish, enforced at the data layer.
Complete audit trail
Every AI decision, override and version freeze is logged with timestamp and actor. Nothing is ever overwritten.
Fully compliant
Fully compliant with DPDPA and global data security and privacy regulations — PII encrypted at rest and in transit.
Accolades to Qdemy
"Six weeks of evaluation camps became six days. Faculty reviewed flags instead of marking piles."
"The co-pilot fixed a broken model answer across 400 already-graded papers in one command. That sold me."
"We calibrated against 40 human-graded papers and the agreement rate has stayed above 91% ever since."
"Students stopped filing re-check requests once they could see exactly where every mark came from."
"It read handwriting our own examiners argued about — and showed its reasoning for the score."
"Anonymized marking with a full audit trail was the requirement. Qdemy was the only platform that had both built in."
Common questions
What is Qdemy and who is it for?▾
An end-to-end examination intelligence platform for universities, boards, and large coaching institutions — rubric design, AI evaluation, on-screen human marking, and results publication for handwritten exams at scale, from a few hundred to 20,000+ students per session.
Does it support both AI and human on-screen marking?▾
Yes — both modes coexist. AI evaluation runs the full 5-stage pipeline in minutes; On-Screen Marking routes the same scanned scripts to human examiners. Use one mode per subject or combine both with a moderation step.
How accurate is the AI evaluation?▾
Over 99% OCR accuracy on standard ruled booklets and AI–human agreement above 91% after calibration. Upload 30–50 human-graded papers and the platform tunes its strictness (γ) to match your examiners’ grading style.
How does it handle partial credit and complex rubrics?▾
Each question is broken into scoring components with their own marks and model answers. The AI scores each independently, reasoning about partial demonstrations of knowledge rather than requiring an exact match.
What happens when the AI flags a low-confidence answer?▾
The evaluation cockpit surfaces it with a warning. An evaluator reviews the AI’s reasoning trace and accepts or overrides the score with a note — every override lands in the audit trail.
Does it support LaTeX, diagrams, and graphs?▾
Natively. LaTeX renders in rubrics and model answers, diagram annotation checks student drawings against labelled references, and graph recognition detects axes, curves and data points for partial marks.
How does anonymization work?▾
Every student gets an Anonymous ID at session start. Evaluators — human or AI — never see names or roll numbers. The mapping is stored separately and revealed only after results are published, enforced at the data layer.
Is it DPDPA compliant?▾
Yes. Built for India’s Digital Personal Data Protection Act: PII encrypted at rest and in transit, scripts never leave the platform unless a COE administrator exports them, and every action carries an audit trail.
How long does onboarding take?▾
Most institutions are live within two weeks: week one covers setup, booklet calibration and a pilot rubric; week two is a dry run on real scripts. After sign-off, sessions run without QverLabs engineers in the loop.
Start grading smarter
A 30-minute demo: we walk through your current workflow, show the platform live, and scope setup for your institution. Most are live within two weeks.