Modular Inverses is the habit of dividing in modular arithmetic by multiplying by an inverse when it exists. In contest math, that habit turns a crowded setup into a relation the student can test, bound, count, or compute. MathGrit teaches it as a recognizable signal, a deliberate move, and a final translation back to the original question.
Solve the congruence $14x \equiv 10 \pmod{18}$ for all $x$ modulo $18$.
1
Here $\gcd(14, 18) = 2$, so $14$ is NOT invertible mod $18$ — but $2 \mid 10$, so solutions exist. Divide the whole congruence (including the modulus) by $2$: $7x \equiv 5 \pmod 9$.
Lift back to modulus $18$: $x \equiv 2 \pmod 9$ gives two residues $x \equiv 2$ and $x \equiv 11 \pmod{18}$. Check $14 \cdot 2 = 28 \equiv 10$ and $14 \cdot 11 = 154 \equiv 10 \pmod{18}$. ✓
Answer:
$x \equiv 2 \pmod 9$, i.e. $x \equiv 2$ or $11 \pmod{18}$
Olympiad / Challenge
Let $p > 2$ be prime. Prove that $1^{-1} + 2^{-1} + \cdots + (p-1)^{-1} \equiv 0 \pmod p$, where each $k^{-1}$ is the inverse of $k$ modulo $p$.
1
Since $p$ is prime, every $k \in \{1, \dots, p-1\}$ is invertible mod $p$, and the map $k \mapsto k^{-1}$ is a bijection of $\{1, \dots, p-1\}$ to itself (its own inverse is $k \mapsto k$).
2
Therefore the multiset $\{1^{-1}, 2^{-1}, \dots, (p-1)^{-1}\}$ is just a permutation of $\{1, 2, \dots, p-1\}$ modulo $p$.
3
Summing, $\sum_{k=1}^{p-1} k^{-1} \equiv \sum_{k=1}^{p-1} k = \frac{(p-1)p}{2} \pmod p$.
4
Because $p > 2$, the integer $\frac{p-1}{2}$ is well defined, so $\frac{(p-1)p}{2} = p \cdot \frac{p-1}{2} \equiv 0 \pmod p$. Hence the sum of inverses is $\equiv 0 \pmod p$.
Answer:
$\sum_{k=1}^{p-1} k^{-1} \equiv 0 \pmod p$ for every prime $p > 2$.
Going Deeper
Generalization: the units mod $n$ — the residues coprime to $n$ — form a group of size $\varphi(n)$, so each has a unique inverse, computable either by the extended Euclidean algorithm (solve $ax + ny = 1$) or by Euler as $a^{-1} \equiv a^{\varphi(n) - 1} \pmod n$. Solving $ax \equiv b \pmod n$ in general: it is solvable iff $g = \gcd(a, n) \mid b$, and then it has exactly $g$ solutions modulo $n$.
Where it appears: solving linear congruences, constructing CRT solutions, the back-substitution in the extended Euclidean algorithm, RSA decryption, and any 'divide both sides by $a$' step in modular algebra.
Pitfall: you may divide by $a$ mod $n$ ONLY when $\gcd(a, n) = 1$. If $a$ shares a factor with $n$, $a$ has no inverse and cancellation can create or destroy solutions — $2x \equiv 2 \pmod 6$ has $x \equiv 1$ AND $x \equiv 4$, not the single solution naive cancellation suggests. When $\gcd(a,n) = g \mid b$, divide $a$, $b$, AND the modulus by $g$ first, then solve and lift back to $g$ solutions mod $n$.
Spot the Signal
Look for problems where the key step is dividing in modular arithmetic by multiplying by an inverse when it exists.
You can describe the hard part as dividing in modular arithmetic by multiplying by an inverse when it exists, but a direct attack starts producing clutter.
The problem rewards preserving structure instead of expanding, listing, or guessing too early.
Learn the Move
Start by identify the integer constraint that calls for modular inverses, then rewrite the givens around it.
Name the relation that makes Modular Inverses legal before doing computation.
Use the new relation to replace the messiest part of the problem with a cleaner one.
Translate the result back to the quantity the problem actually asks for.
Avoid These Traps
Do not use Modular Inverses just because the surface looks familiar; verify the required condition first.
Applying Modular Inverses because it sounds relevant, without checking the trigger first.
Stopping after spotting the technique instead of finishing the calculation or proof.
MathGrit Coach Note
Let modular inverses reveal the integer structure; then compute only what remains.
Try it on:
Practice a contest problem where the key step is dividing in modular arithmetic by multiplying by an inverse when it exists.