LeetCode Algorithms and Data Structures: The List That Covers Most Problems

Most LeetCode problems use about ten algorithm families, not hundreds of named algorithms. Hashing, two pointers, sliding window, binary search, sorting, depth first search, breadth first search, heaps and dynamic programming cover the large majority.

The names matter less than the recognition. LeetCode almost never asks you to write a named algorithm. It describes a situation, and you have to see which family fits.

Learn the data structures first. Every algorithm below runs on an array, a string, a hash map, a tree, a graph or a heap.

The ten families that appear most

Family	What it does	Example problem
Hashing	Stores and looks up a key in constant average time	Two Sum
Two pointers	Moves two indexes through one sorted array	Container With Most Water
Sliding window	Grows and shrinks a range over an array	Longest Substring Without Repeating Characters
Binary search	Cuts a sorted range in half each step	Search in Rotated Sorted Array
Sorting	Orders the data so a single scan solves it	Merge Intervals
Depth first search	Follows one path to the end, then backs up	Number of Islands
Breadth first search	Visits nodes level by level	Word Ladder
Heap	Keeps the k largest or smallest items	Top K Frequent Elements
Dynamic programming	Reuses answers to smaller cases	Coin Change
Backtracking	Builds a partial answer and undoes bad choices	N-Queens

If you know these ten well, most medium problems become readable. You will still have to think, but you will know where to start.

The data structures you need first

An algorithm is only a way to move through a data structure. So learn the structures before the tricks.

Arrays and strings. Almost every problem starts here.
Hash map and hash set. These turn many O(n squared) scans into O(n).
Stack and queue. A stack drives depth first search, a queue drives breadth first search.
Linked list. Mostly pointer work, such as reversing or finding a cycle.
Binary tree. Inorder, preorder and postorder traversal are the base skills.
Graph. Store it as an adjacency list, then run depth first or breadth first search.
Heap. A structure that gives you the smallest or largest item in O(log n).

Our data structures course for coding interviews walks through each one with code.

Algorithms that show up less often

These are worth learning after the ten above, not before.

Union find. Also called disjoint set. It appears in graph grouping and cycle problems.
Topological sort. Orders tasks that depend on each other, as in Course Schedule.
Dijkstra's algorithm. Shortest path when edges have different weights.
Trie. A prefix tree for word search and autocomplete questions.
Bit manipulation. XOR tricks and bitmasks for a small set of problems.

Segment trees, Fenwick trees, KMP and Rabin-Karp are real algorithms, but they are rare in interviews. Skip them until the common set is solid.

What interviewers ask for on top of the algorithm

Naming the right algorithm is only half the answer. You also have to state the cost and defend it.

For every problem, say the time and space complexity out loud. Then say why you chose that structure over another one. A correct solution with no complexity analysis reads as an incomplete answer.

Sorting and searching come up in almost every round, so start there. We cover the ranking in top algorithms for sorting and searching in interviews.

How to Prepare

Learn patterns, not problem counts. Grouping problems by shape beats grinding a random list. Grokking the Coding Interview organizes questions into the patterns above.
Start with a small, ranked set. Grokking the 75 Top Coding Interview Questions covers each family once, which is enough for most screens.
Solve five problems per family before moving on. Recognition comes from repetition inside one family, not from variety.
Write the complexity before you write the code. If you cannot state the target cost, you have not chosen an approach yet.
Redo the problems you failed after a week. A second attempt shows whether you learned the pattern or memorized the answer.
Read the common pattern list. What are the most common LeetCode patterns? maps problem wording to the family it belongs to.
Decide how much LeetCode you need. Is using LeetCode worth it? sets out where the platform helps and where it does not.