225 questions. 20 units.
One learning intelligence system.
The existing KARMA PYQ production engine is more than a question collection: official source → year/shift/QID → original solution → independent QA → cross-chapter concepts → misconception diagnosis → test/revision → manuscript → publishing. This public projection reveals aggregate intelligence only; question wording, unreleased original solutions and student data are not exposed.
Chapter frequencies
Read-only canonical primary-unit counts. They sum to 225 sampled questions; not predictions for 2027. Select a unit to inspect its reviewed-source coverage.
Select a chapter
Question-level canonical pipeline
| Stage | Record / QA requirement | Current evidence |
|---|---|---|
| 01 · Source | Exam → year → session/date/shift → official paper/QID → official key | 225 identities in the owner ledger |
| 02 · Rights | Allowed source reference vs prohibited commercial question reproduction | Original NTA wording/images: HOLD |
| 03 · Canonicalize | Chapter/topic/subtopic, linked units, duplicate/variant groups | 20 canonical units; 14 aliases/cross mappings previously harmonized |
| 04 · Solve | Original reasoning, equations, alternative methods, shortest method | 222 internal QA passes; 2 defects closed; 1 dispute remains |
| 05 · Explain | Difficulty, concept tested, common trap, time/strategy, exam overlay | 172 difficulty bands; 53 uncalibrated |
| 06 · QA | Independent reviewer → official key → visual/equation/rights check | 30 source-schema row misalignments require source-native reconciliation |
| 07 · Deliver | BalaJEE assessment/diagnostic/revision and VYOM teaching derivative | Existing KARMA knowledge, BalaJEE catalog and assessment systems—not a duplicate store |
| 08 · Publish | Interior, cover, ISBN, print proof, digital entitlement, marketplace, errata | First real book FINAL_PUBLICATION_PASS remains OPEN |
Evidence-based learning route
Concept tag → misconception → prerequisite → original practice → learner response → diagnostic → targeted revision → spaced retrieval → verified next best action.
Unauthenticated browsing cannot change learner mastery or publish a private test attempt.
Open existing BalaJEE learner journey ↗Original diagnostic example
At fixed temperature, a reaction has Q=2 and K=10. Which net direction is thermodynamically favoured?
Reveal worked reasoning
Because Q is below K, the net change is forward until equilibrium is reached, assuming reaction is possible. This does not imply any rate. This is an independently authored example, not official NTA question text.
If a learner confuses equilibrium with rate, a follow-up must test both reaction quotient and kinetic concepts rather than awarding mastery for opening an explanation.
Nine planned books, three bundles
Blocker Theory
B-PYQ-SCHEMA: 30 original rows have J/K/L field-layout drift. Source untouched. Reconcile official QIDs and original owned notes before treating fields as standardized.
B-PYQ-RIGHTS: commercial reproduction of protected third-party text/figures not cleared. Continue source references, original problem authoring and rights-safe analysis.
B-PYQ-Q64: P09 Q64 disputed key/expert decision; preserve record and proceed with the other independent 224 records.
B-PYQ-DIFF: 53 unrated source records; do not invent difficulty.
Independent acceptance gates
Official syllabus → source identity → authorized input → canonical classification → independent answer QA → alternate method/trap QA → student learning evidence → interior proof → publisher/rights → physical print proof → genuine order → fulfillment → bank settlement → errata.