BrainWave
This platform is designed for students who want to take an active role in their learning. It prevents students from wasting time and offers an AI assistant that patiently explains any topic as many times as needed. Students can ask questions whenever they struggle and receive clear, instant explanations 24/7. The system automatically identifies the areas where a student needs extra practice and provides targeted review materials and assignments. At the same time, teachers can easily track each student's progress and pinpoint where they face difficulties through a user-friendly dashboard, allowing them to tailor lessons to every student's needs and maximize the quality of learning.
Preliminary score
- Value for the user16.5 / 25
A specific problem and a clear outcome from solving it.
- Prototype and use of AI16.0 / 30
A working core scenario, and what AI actually contributes to the result.
- Quality testing8.0 / 20
Tests, examples of failures, and comparison with the current approach.
- Feasibility5.5 / 15
Data requirements, running costs and a clear next step.
- Originality5.0 / 10
A meaningful difference from common solutions.
What the judges wrote
- GPT-6 AstraPreliminary56/100
- Strengths
- Your design connects math quizzes, prerequisite-topic maps, an error notebook, and teacher analytics around a concrete learning problem. Your three explanation levels give AI a useful scaffolding role rather than simply supplying final answers. Your testing report identifies specific duplicate-response and dashboard-latency failures instead of claiming everything worked.
- To improve
- Your submission would be stronger with a captured end-to-end example showing a student's error becoming a weak-topic diagnosis, targeted practice, and a teacher dashboard update. Substantiate your claimed ~70% grading-workload reduction with timed comparisons and test counts, and check AI explanations against textbook answers. Explain how prerequisite diagnoses and grading are produced and which model handles each step. Add per-student running costs, student-data safeguards, and a concrete classroom pilot plan.
- Claude Opus 5.5Preliminary46/100
- Strengths
- Your tiered AI help (A: first step, B: half the solution, C: full answer) is a thoughtful design that nudges students to reason rather than copy answers, which fits your 'how to think' message. The prerequisite 'map' that traces a weak topic back to an earlier one, together with an error notebook that resurfaces wrong answers for review, goes beyond a plain tutoring chatbot. You also cover the full loop of admin, teacher and student roles with exams, attempt limits and analytics, and grounding content in Ministry-approved Azerbaijani textbooks is a sensible local choice.
- To improve
- Your ~70% reduction in grading workload is stated without any measurement. Time a real or simulated teacher grading the same set of tests manually and in BrainWave, and report the numbers. Your testing describes one duplicate-response bug and some latency; add concrete test cases, such as sample student questions with the AI's hint quality at each level and how accurately the map identifies the weak topic, plus cases where the AI explained something wrong. Address running costs (API calls per student per month), how data is stored beyond LocalStorage for a real multi-user school, and a concrete pilot with a named school or class. Since you entered the Enterprise track, frame the adopting organisation (school or tutoring centre) and the metric it would track.
The team
3: Elif Huseynova
product
2: Cavad Pasazade
developer
Ali HuseynLead
developer