ISI and CMI Entrance: Join the Next Batch

The ISI and CMI entrance exams are not harder versions of JEE. They test something different, and students who prepare for them as though they were board exams with tougher sums almost always underperform.

Our ISI and CMI course is built around that difference. The next batch is enrolling now.

The difference that decides results

JEE rewards speed and pattern recognition across a wide syllabus. ISI and CMI reward proof.

The subjective papers ask you to justify a claim, not produce a number. A correct answer with no valid argument earns very little. An elegant argument reaching a small result earns a lot.

That is a genuinely different skill, and it is the one most entrance coaching never teaches because it cannot be marked by an answer key.

The structure you are preparing for

Both follow a two-stage shape: an objective first paper that screens, then a subjective paper that decides.

For ISI (B.Stat and B.Math), the first paper is multiple-choice and the second requires full written solutions.

For CMI (BSc Mathematics and Computer Science), the paper mixes objective questions with problems requiring complete arguments.

The screening paper is a filter you must clear efficiently. The subjective paper is where selection actually happens. Preparation weighted entirely toward the first is the standard mistake.

What the syllabus really covers

The named syllabus is close to higher secondary mathematics: algebra, inequalities, functions, sequences and series, combinatorics, number theory, coordinate and plane geometry, trigonometry, and single-variable calculus.

The depth is not. Questions combine two or three areas and reward an idea rather than a procedure. Olympiad-style thinking is directly useful here, which is why strong IOQM and INMO candidates often do well.

Combinatorics and number theory deserve particular attention: they appear heavily and are the areas school courses cover most thinly.

How we teach proof

Proof is taught explicitly, not absorbed by osmosis.

Induction, contradiction, contrapositive, pigeonhole, extremal arguments, invariants, and constructing an explicit counterexample — each introduced on problems where it is the natural tool, then mixed so you must choose.

Written solutions are marked the way the real papers are marked. Students are consistently surprised by this early on: work they considered complete loses marks for gaps they could not see. Learning to spot your own unjustified step is most of the improvement.

What a marked solution actually looks like

This is the part students find hardest to believe until they see it.

Take a typical subjective problem: prove that among any five integers, some two have a difference divisible by four. The answer is pigeonhole — four possible remainders, five integers, so two share a remainder and their difference is divisible by four.

A student who writes only the final sentence gets little. A student who states the four residue classes, observes that five objects into four classes forces a repeat, and then concludes gets full marks. Same insight, very different scores.

The marks live in the justification, and that is a writing skill as much as a mathematical one. It is also why one-to-one feedback on written work is the only thing that reliably improves it.

The topics that decide outcomes

Combinatorics appears constantly and school courses barely touch it beyond basic permutations. Counting arguments, bijections, inclusion-exclusion and pigeonhole are all standard here.

Number theory is similar: divisibility, modular arithmetic, primes and the structure of integers. Almost no school syllabus develops it, and ISI leans on it heavily.

Inequalities reward technique: AM-GM, Cauchy-Schwarz, and knowing when to substitute rather than expand.

Functional equations and sequences appear regularly and reward the habit of testing small cases before attempting a general argument.

Calculus appears but rarely decides outcomes. Students who spend their preparation on calculus because it feels like the advanced topic are optimising the wrong thing.

ISI, CMI, or both?

Most serious candidates sit both, and the preparation overlaps heavily enough that this costs little extra.

ISI's B.Stat and B.Math are statistics and mathematics respectively, with the institute's research culture behind them. CMI's BSc programmes in Mathematics and Computer Science suit students who want the computing side alongside the mathematics.

The exams differ in format more than in the thinking they demand, so preparing for the harder proof requirements of ISI's subjective paper leaves you well placed for CMI. Sit both unless there is a specific reason not to.

How the screening paper should be handled

The objective paper is not where selection happens, but it is where candidates get eliminated, which makes it worth a specific strategy rather than general confidence.

It rewards the same thinking as the subjective paper, compressed. Many questions resolve by testing a small case, checking a boundary, or eliminating options that fail an obvious constraint. Full derivations are usually unnecessary and often too slow.

Practise it timed and separately from proof work. Students who prepare only through long-form problems can be genuinely strong mathematically and still miss the cut on pace, which is an avoidable way to lose a year.

Timeline and the batch

Serious preparation takes nine to twelve months alongside Class 11 and 12. Starting in Class 11 is ideal; a focused Class 12 run is possible but tight, since proof-writing improves with accumulated feedback rather than intensity.

Classes are online in small groups, because every written solution needs individual comment to be worth anything. Batches are capped and close in order of enquiry.

We will mark one proof-based problem set before you commit and show you exactly how it would score. That is the most honest preview of the work available.