VIDHYASETU · AI-ASSISTED EVALUATION
Give teachers a first evaluation they can inspect, correct and approve. Turn handwritten answer sheets into question-wise marks, with your school’s answer key at the centre.
For principals, academic heads and school management.
x² − 5x + 6 = 0
Crossed out: x² − 2x + 3x + 6 = 0
Corrected: x² − 2x − 3x + 6 = 0
x(x − 2) − 3(x − 2) = 0
(x − 2)(x − 3) = 0
x = 3; the second root is missing.
The student corrects a sign mistake and reaches the correct factors. One root is missing, so one of the two marks for roots is withheld.
Inspect the answer. Review the reasoning. Correct and approve.
Fictional handwritten samples explain the review process. This is not a product screenshot or an accuracy test.
01 / THE SCHOOL’S PRIORITIES
A useful evaluation system must help the people running the exam. Here is what each team should look for in VidhyaSetu.
Follow processing, review and results across the exam workflow. Judge value using time to approved results and total staff effort.
Can your team complete the exam cycle with less effort and clear responsibility?
Review the answer, suggested marks and reasoning together. Correct partial credit, alternative methods and misread handwriting before approval.
Can teachers explain and stand behind the marks they approve?
Organise the answer key and scanned papers, process a ZIP batch and follow its progress while assigned teachers handle academic review.
Does the hand-off from scans to teacher review work for your team?
02 / FROM PAPER TO APPROVED MARKS
Students continue writing on paper. Your team digitises the answer sheets for evaluation and keeps the final academic decision.
Create the exam, organise classes and subjects, and assign teachers. Agree who owns review and final approval.
Upload the answer key and marking criteria, then the scanned answer-sheet batch. Check page order, completeness and legibility.
Process papers against the answer key. Track the batch and inspect extracted answers, suggested marks and question-level reasoning.
Check the original paper, correct suggestions and approve marks. Use the results view to follow the class’s evaluation progress.
Include maximum marks, required steps, partial-credit rules and acceptable alternatives. A clearer answer key gives teachers a stronger basis for reviewing the first evaluation.
03 / IS IT RIGHT FOR YOUR SCHOOL?
A focused first use case makes it easier to judge both the academic quality and the operational benefit.
Confirm the pilot’s class, subject, language and question formats before using it for a live exam.
04 / VALUE IN YOUR SCHOOL’S TERMS
Replace generic speed claims with your own numbers. Count teacher review and the work of scanning and setup, then validate the estimate in a pilot.
Time released can support teaching, moderation and student feedback. It is not automatically a cash saving.
Example values are editable assumptions, not measured school results. Formula: papers × (current minutes − review minutes) ÷ 60 − extra setup hours. Include rescanning, corrections and moderation in your estimates. This compares staff effort; AI processing time, queue delays, subscription charges and deployment costs are excluded. Validate turnaround and total cost separately in the pilot.
05 / PROVE IT BEFORE YOU EXPAND
Choose one class, one subject and a representative batch. Agree what success means before processing begins.
YOUR PILOT SCORECARD
Agree acceptable thresholds with your school, then use the results to decide the next exam cycle.
06 / ACCOUNTABILITY BY DESIGN
VidhyaSetu is part of IgniteX’s governed AI approach: AI assists the work, and accountable people retain the decision.
Explore IgniteX’s governance approachUse review, correction and approval as part of the exam process. Resolve doubtful answers before finalising marks.
Confirm access, hosting, external processing, retention, deletion and data-use terms for the selected deployment.
Use synthetic or school-authorised anonymised samples for the first demo. Keep names, roll numbers and other identifiers out of the enquiry form.
07 / BEFORE YOU DECIDE
Academic fit, student data and commercial terms belong in the same conversation.
Share approximate paper volume, pages per paper, exam frequency and support needs. Ask for a complete exam-cycle cost.
No. It prepares a first evaluation using the answer key and marking criteria. Teachers inspect the original answer, review the suggested marks and reasoning, make corrections and approve the final result. Your school remains responsible for academic decisions.
The current workflow processes scanned handwritten answer sheets and suggests question-wise marks. Suitability depends on legibility, scan quality, answer length and the question format. Include descriptive answers, equations, diagrams and difficult handwriting in your demo sample; confirm support for each required format before a pilot.
Start with the exact class, subject, language and paper format your school uses. We will demonstrate that combination and agree the pilot scope. We do not claim universal support across every board, script, subject or examination pattern.
The answer key should define step marks, acceptable alternatives and partial-credit rules. Review those criteria before processing a batch. Teachers can inspect and correct the suggested marks when an answer uses a valid alternative method or the first evaluation misses context.
Compare the extracted answer with the original paper and correct the evaluation where needed. If the source is unreadable or a page is missing, arrange a better scan or manual checking. Include these cases in the pilot; an AI suggestion should never settle a doubtful mark on its own.
There is no published school-wide accuracy percentage or guaranteed papers-per-hour rate on this page. Measure question-level agreement with your teachers, material marking errors, correction time and total time to approved results on a representative batch. Page count, question complexity, scan quality and deployment capacity affect the outcome.
Use synthetic papers or school-authorised anonymised samples for the first demo. Before any live pilot, agree the hosting location, who can access papers, whether external model providers receive data, retention and deletion terms, and how data may be used. Confirm these details for the proposed deployment before sharing identifiable student records.
Begin with a bounded evaluation workflow alongside your current process. If you need marks exported or synchronised with an ERP, agree the required file format or integration, owner, cost and acceptance test in the proposal. Do not assume a ready-made connector to every school system.
A quote depends on paper volume, pages per paper, exam frequency, deployment capacity, onboarding, support and any integrations. Ask for the total cost for your exam cycle and a clear breakdown of setup, recurring charges, usage limits and support. The workload calculator estimates staff effort only; it is not a price quote.
Start with one class, one subject and one representative batch. Agree the marking scheme, school reviewer and acceptance criteria first. Compare against teacher-checked reference papers and decide whether to expand after reviewing both marking quality and staff effort. Pilot scope and commercial terms are agreed before it begins.
YOUR PAPERS. YOUR CRITERIA. YOUR DECISION.
Walk through the answer key, first evaluation, teacher correction and result. Leave with a clearer view of fit, pilot scope and the evidence your school needs.