AI learning for school students · STEM first

A learning app that remembers what each student is weak at.

Vedora is being built as a personalised study workspace for Classes 7-12. We are starting with NCERT-aligned mathematics and expanding into physics, chemistry, and biology, using each student's weak-concept history to shape explanations, practice, revision, and weekly tests.

Math Physics Chemistry Biology

See why it is different

Goal 72%

Streak 6d

Focus 3

Weak concept detected

Balancing equations is slowing down chemistry accuracy.

Vedora adjusts today's practice and this week's test around the exact concept gaps that keep repeating.

Algebra signs Chemical equations Force diagrams

AI answer style

Explain with one simpler example first. Student has missed this concept twice this week.

Weekly repair map auto-built

7-12 school years in focus

4 STEM subjects planned

1 personal concept memory per student

What we are building

A student workspace that connects learning, practice, revision, and testing.

Learn from the exact gap

Lessons and hints are selected from the student's weak concepts instead of showing the same generic explanation to everyone.

Practice that adapts

Every attempt updates a concept profile, so the next question can target the reason behind the mistake, not only the chapter name.

Weekly tests from real weakness

Instead of random mocks, Vedora can generate weekly tests around the weak areas a student actually showed during the week.

Reports that lead to action

Students, parents, and teachers see what to fix next: concept gaps, repeated errors, effort, accuracy, and revision priority.

Why Vedora is novel

Most apps remember scores. Vedora remembers concepts.

A score tells you that a student got 6 out of 10. It does not tell you whether they struggle with signs, units, force diagrams, mole ratios, graph reading, or exam pressure. Vedora is built around a persistent weak-concept memory, so every answer, hint, revision card, and test can become more personal over time.

Weak-concept graph Every mistake is mapped to concepts and error patterns, creating a living profile of what the student needs next.

Personalised answers The AI can explain the same question differently for different students because it knows their recent weak spots.

Revision timing Wrong notes and reviews can be scheduled around forgetting risk, repeated errors, and test deadlines.

Adaptive weekly tests Weekly tests are generated from the student's actual learning trail, so testing becomes diagnosis plus repair.

Building logic

The product loop gets smarter as the student studies.

Student attempts a question

Vedora captures the answer, confidence, step, and mistake pattern.

Concept memory updates

The system strengthens or weakens concept signals for that student.

Next action is personalised

The app chooses a hint, lesson, practice set, wrong-note review, or mock.

Weekly test closes the loop

The week's repeated weak concepts become the student's next test plan.

Current status

Pre-beta product, starting with math and designed for STEM expansion.

Math learning flow is being built first with lessons, practice, notes, mocks, and reports.
The same concept-memory model is intended to support physics, chemistry, and biology later.
Content is structured and reviewed so AI helps personalise learning without becoming an unchecked answer machine.

For students, parents, teachers, and early testers