Generated by Codex with GPT-5
HARDArray, String, SimulationThe input is a list of words and a fixed line width. The task is to break the words into lines and return the exact text layout that a formatter would produce.
Continue ...Generated by Codex with GPT-5
MEDIUMLinked List, Recursion, Iteration, Pointer ManipulationThe input is the head of a singly linked list. Every adjacent pair of nodes must be swapped:
1 -> 2 -> 3 -> 4 becomes 2 -> 1 -> 4 -> 31 -> 2 -> 3 becomes 2 -> 1 -> 3The important constraint is that node values cannot be copied around. The solution has to rewire the list itself.
Continue ...Generated by Codex with GPT-5
MEDIUMArray, Two Pointers, Binary SearchThe input has length n + 1, but every value is guaranteed to be in the range 1 through n.
That guarantee means one value must appear more than once. So the duplicate is not the hard part. The hard part is satisfying the interview constraints at the same time:
Continue ...Generated by Codex with GPT-5
MEDIUMGraph, DFS, BFS, Union-FindEach equation like a / b = 2.0 gives a multiplicative relationship between two variables.
The query x / y asks:
can the known equations connect x to y, and if they can, what ratio do those equations imply?
Generated by Codex with GPT-5
MEDIUMMath, RecursionThe task is to compute x^n, but the real constraint is not the arithmetic itself. It is that n can be very large, including negative values, so a straightforward loop that multiplies x by itself |n| times is too slow.
Generated by Codex with GPT-5
HARDLinked List, RecursionThe list must be processed in consecutive chunks of size k.
If a full chunk exists, those k nodes should be reversed in place. If fewer than k nodes remain at the end, that tail must stay exactly as it is.
Generated by Codex with GPT-5
MEDIUMArray, Bit ManipulationThe array contains one value that appears exactly once, while every other value appears exactly three times. The obvious Counter or hash-map solution works, but it uses extra memory, so it does not satisfy the strongest version of the problem.
Generated by Codex with GPT-5
MEDIUMArray, Dynamic Programming, GreedyEach position tells how far forward it can jump. The goal is not to maximize distance on each move. The goal is to reach the last index using as few jumps as possible.
Continue ...Generated by Codex with GPT-5
MEDIUMArrayThe input intervals are already sorted by start time, and they do not overlap with each other.
That matters because the problem is not asking to rebuild the whole interval set from scratch. It is asking where one new interval fits into an already clean timeline.
Continue ...Generated by Codex with GPT-5
MEDIUMArray, Hash Table, MatrixThe rule sounds simple:
if any cell contains 0, then every cell in that cell’s row and every cell in that cell’s column must become 0.