In many courses and educational programs, one instructor is responsible for too many students, so learners cannot receive detailed, actionable feedback on their solutions quickly enough. As a result, students either wait a long time for manual grading or stop iterating on their work, even though revisions and error analysis are critical for learning.
Existing automated checkers (e.g., typical online judges and LMS systems) often provide only a binary “correct / incorrect” result and do not explain what exactly is wrong, how to fix it, or how to approach the task differently. At the same time, teachers spend a large share of their time on repetitive grading instead of teaching activities such as lectures, seminars, and individual support.
Report
We are building AiChecker: an AI-powered system that checks student solutions and provides repeatable, detailed feedback. It highlights specific mistakes, gives corrective hints, and answers clarifying questions about the task statement. The goal is to improve learning outcomes by enabling fast iteration for students and to reduce grading workload for teachers.
Our core hypothesis is that if we introduce an AI system that analyzes solutions, points out concrete errors, provides targeted hints, and supports Q&A on the problem statement, then students will revise solutions more often and understand mistakes better, while teachers will spend significantly less time on grading, because feedback becomes fast, actionable, and scalable.
Target users:
- Educational organizations and teachers
- Students
Current status (MVP):
We already have a working MVP that is actively tested by school students and math clubs. We continuously collect feedback from teachers and iterate on the product. We aim to improve the system, strengthen validation with metrics, and explore partnerships and investment opportunities.
Link to our platform: ai-checker-phi.vercel.app
