HSC is two years of syllabus, and almost nobody gets through the first year without skipping something. Maybe it was vectors in physics 1st paper, or the mole calculations in chemistry, or a month of higher math lost to a family wedding. It did not seem to matter at the time. Then the 2nd paper arrived and started building on it.
That is a very common shape of trouble in HSC preparation: not a weak subject, but one skipped chapter that everything later leans on. This guide is about using AI to go back and rebuild it — properly, from the prerequisite up — without losing three weeks to it.
Why don't HSC chapters stand alone?
HSC science is a dependency graph pretending to be a list of chapters. A few of its edges:
- Physics: vectors sit under almost all of mechanics; waves in the 1st paper sit under physical optics in the 2nd; calculus from higher math sits under both.
- Chemistry: the mole concept sits under every quantitative question; equilibrium sits under acids, bases and pH; bonding and electronegativity sit under most of organic chemistry.
- Higher math: functions and limits sit under differentiation, which sits under integration — and under half of physics.
Skip a node and the chapters above it do not become harder. They become impossible to understand and only possible to memorise, which is a much worse kind of hard.
Step 1: start where you actually are
Name the chapter — say, physical optics — and Porphy asks a few short questions before it researches anything. It does not ask you to rate yourself, because nobody knows whether they are 'intermediate' at optics. It asks about concrete things instead: can you say what a wavelength is, have you met superposition, what are you studying for. Those answers change the research plan, so the course starts at your gap rather than at page one.
Pick the depth to match the time you have: a 3–4 part overview for a chapter you half-know, a 5–7 part standard course for one you skipped outright. Every part is researched on the live web and ships with the sources it was written from — here is what happens during that run.
Step 2: read with the tutor beside you
Say the course reaches Young's double-slit experiment, and the line you stall on is this one: dark fringes form where the path difference is half a wavelength (or any odd number of half-wavelengths). You can memorise that. Or you can highlight it, press the mic, and have the voice tutor explain it out loud while it draws two waves on a whiteboard — one shifted by half a wavelength, crest landing on trough, the two cancelling to a flat line.
In Bangla it sounds like a good teacher at the board, with the technical terms left in English:
path difference যখন wavelength-এর অর্ধেক, তখন একটা wave-এর crest পড়ে আরেকটার trough-এর উপর — দুটো মিলে cancel হয়ে যায়, তাই ওখানে dark band।
Once that sentence is understood rather than memorised, the fringe-width formula stops being something to remember and becomes something you could derive in the exam hall.
Step 3: find out which kind of 'don't know' you have
After a skipped chapter, 'I don't know it' can mean four different things, and they need different fixes. Every concept Porphy teaches is scored on four axes — memory, comprehension, structure and application — and an evaluator agent updates them after each MCQ set, flashcard run or explain-it-back attempt with the tutor.
Organic chemistry is the classic case. Plenty of students can list every named reaction (memory: high) and still cannot predict the product of one they have not seen (application: low). Rereading fixes the first and does nothing for the second. The knowledge map tells you which one you have, and links each weak concept straight back to the paragraph that teaches it.
Step 4: bring your own lecture sheets and past papers
Your college teacher's sheet, a chapter of your textbook, a past paper — if it is a typed PDF, upload it and every tool works on it the same way: highlight a line for the tutor, generate MCQs from it, map its concepts, ask questions about it. A photographed page will not work. There is no OCR step, so the PDF has to contain selectable text.
From HSC to the admission test
Admission tests after HSC reuse the same syllabus and ask it differently: faster, less predictably, and far more often at the application end. A chapter you memorised for the board exam is the one that lets you down there. That is the practical case for rebuilding a skipped chapter properly now instead of patching it for one exam — you will be tested on it again within months.
What AI will not do for your HSC
- It does not know your board's past questions or what your teacher expects. Use it alongside your textbook and past papers, not instead of them.
- It can be wrong. Check any line that surprises you against the sources each part lists, and against the book.
- It will not do the problem practice. Numerical physics and chemistry marks come from solving problems by hand, against the clock.
- It is an AI, not a private tutor. It is there at 1am and will explain the same line five ways without sighing, but it will not notice that you have stopped trying. Where a human tutor is still better is worth reading honestly.
An HSC preparation plan for one skipped chapter
- Name the chapter and answer the intake honestly — your answers decide where the course starts.
- Read the first part. Highlight every line you cannot explain, and take the worst one to the voice tutor.
- Explain the core idea back to the tutor, and fix what it says you left out.
- Run one MCQ set and one flashcard deck on the course.
- Open the map. Tomorrow's first twenty minutes go to the concept with the lowest application score.
If you are a level earlier, the same method works there too — here is the SSC physics and higher math version.
Every new account starts with 2,000 credits free — no card, no subscription, and they never expire. A full researched course runs about 600–1,200 credits, an MCQ set or flashcard deck 10–30, and the voice tutor 20–60 a minute. Free accounts run a lighter pipeline so the credits go further: at most 3 parts per course, with text and tables in place of generated visualizations.