{"slug": "exam-rescue-from-exam-chaos-to-an-exam-ready-plan", "title": "Exam Rescue — From Exam Chaos to an Exam-Ready Plan", "summary": "A developer built Exam Rescue, a browser-based study planning tool that turns syllabus, study material and previous-year questions into a prioritized day-by-day exam plan. The app pairs client-side planning logic with an AI Study Coach powered by Hugging Face Transformers.js and the open-weight SmolLM2-135M-Instruct-ONNX-MHA model, which downloads on first use and runs inference entirely in the browser via ONNX rather than a proprietary API backend.", "body_md": "*This is a submission for the [Hacktoberfest Weekend Challenge: Build for a Friend](https://dev.to/challenges/hacktoberfest-weekend-2026-10-01)*\n\nI built **Exam Rescue**, a student-focused study planning tool for my roommate.\n\nI built it because she often starts panicking when exams are close and doesn't know where to start when there is a lot of syllabus and study material to cover.\n\nExam Rescue turns that exam-time chaos into a structured workflow.\n\nStudents can enter:\n\nThe app then analyzes the inputs, identifies important topics from the PYQs, matches topics with study materials, assigns priorities, and generates a day-by-day study plan.\n\nIt also includes an **AI Study Coach** that gives additional study priorities based on the student's subject, syllabus and PYQs.\n\nThe workflow is:\n\n**Input → Analyze → Prioritize → Study Plan → AI Guidance**\n\n**Live Demo:** [https://exam-rescue-omega.vercel.app/](https://exam-rescue-omega.vercel.app/)\n\nThe project runs directly in the browser. The AI Study Coach downloads its model when it is used for the first time and performs inference in the browser.\n\n**GitHub Repository:** [https://github.com/priyaupadhyay-2311/Exam-rescue](https://github.com/priyaupadhyay-2311/Exam-rescue)\n\nThe project is built as a frontend application using HTML, CSS and JavaScript.\n\nI built Exam Rescue with HTML, CSS and JavaScript, with the study-planning logic running directly in the browser.\n\nFor the AI component, I used **Hugging Face Transformers.js** with the open-weight **SmolLM2-135M-Instruct-ONNX-MHA** model.\n\nThe model runs in the browser through ONNX instead of requiring a separate proprietary AI API backend.\n\nThe AI Study Coach uses the student's subject, syllabus and previous year questions to generate short study priorities and advice.\n\nThis makes the AI component lightweight and keeps the inference experience directly in the browser.\n\nOpen AI technology made it possible to add an AI Study Coach without building the project around a paid proprietary AI API.\n\nBecause inference runs in the browser, the project does not need to send every AI prompt to a proprietary AI service.\n\nIt also gives the project flexibility: the open model can potentially be replaced or upgraded without redesigning the entire application around a closed provider.\n\nFor a student-focused project, open technologies make experimenting with AI more accessible to a student developer and make it easier to learn how AI can be integrated into real applications.", "url": "https://wpnews.pro/news/exam-rescue-from-exam-chaos-to-an-exam-ready-plan", "canonical_source": "https://dev.to/priya_d350bb33710728f955d/exam-rescue-from-exam-chaos-to-an-exam-ready-plan-4an6", "published_at": "2026-10-04 17:02:36+00:00", "updated_at": "2026-10-04 17:13:02.991619+00:00", "lang": "en", "topics": ["ai-tools", "ai-products", "large-language-models", "developer-tools"], "entities": ["Exam Rescue", "Hugging Face", "Transformers.js", "SmolLM2-135M-Instruct-ONNX-MHA", "ONNX", "Hacktoberfest", "Vercel"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/exam-rescue-from-exam-chaos-to-an-exam-ready-plan", "markdown": "https://wpnews.pro/news/exam-rescue-from-exam-chaos-to-an-exam-ready-plan.md", "text": "https://wpnews.pro/news/exam-rescue-from-exam-chaos-to-an-exam-ready-plan.txt", "jsonld": "https://wpnews.pro/news/exam-rescue-from-exam-chaos-to-an-exam-ready-plan.jsonld"}}