Matrix Questions at Flipkart: What to Expect

Matrix questions appear in roughly 14% of Flipkart's technical interview problems. For a company managing massive logistics, inventory grids, and recommendation systems, matrices are a direct representation of real-world data structures. Your ability to traverse, transform, and analyze 2D arrays efficiently is a strong signal of how you might handle data at scale. Expect these problems to test not just your algorithmic skills, but also your capacity to translate a business scenario—like warehouse slot management or user-item preference grids—into clean, optimized code.

What to Expect — Types of Problems

Flipkart's matrix problems typically fall into three categories. First, classical traversal and search: these include problems like searching in a row-wise and column-wise sorted matrix, or spiral traversal, which test your fundamental loop control and index manipulation. Second, dynamic programming on grids: questions like finding unique paths, minimum path sums, or largest square submatrices are common, assessing your ability to break down complex problems into overlapping subproblems. Finally, graph representations: many matrix problems are essentially graph problems in disguise, where each cell is a node and adjacent moves are edges. This includes number of islands, rotten oranges, or shortest paths in a binary maze. Recognizing this underlying pattern is crucial.

How to Prepare — Study Tips with One Code Example

Focus on mastering patterns, not memorizing solutions. Start by ensuring you can flawlessly write BFS/DFS for matrix traversal. Then, practice applying dynamic programming to grids. Always clarify edge cases like empty matrices or single-row inputs. A key pattern is modifying the matrix in-place to mark visited cells, often by temporarily changing values to avoid using extra space for a separate visited structure.

Here is a classic example: Rotting Oranges (BFS on a matrix). The goal is to find the minimum time for all fresh oranges to rot, where rotten oranges rot adjacent fresh ones each minute.

from collections import deque

def orangesRotting(grid):
    if not grid:
        return 0

    rows, cols = len(grid), len(grid[0])
    queue = deque()
    fresh_count = 0
    minutes = 0

    # Initialize queue with all initially rotten oranges
    for r in range(rows):
        for c in range(cols):
            if grid[r][c] == 2:
                queue.append((r, c))
            elif grid[r][c] == 1:
                fresh_count += 1

    # Directions for 4 adjacent cells
    directions = [(1,0), (-1,0), (0,1), (0,-1)]

    # BFS
    while queue and fresh_count > 0:
        minutes += 1
        for _ in range(len(queue)):
            r, c = queue.popleft()
            for dr, dc in directions:
                nr, nc = r + dr, c + dc
                if 0 <= nr < rows and 0 <= nc < cols and grid[nr][nc] == 1:
                    grid[nr][nc] = 2  # Mark as rotten
                    fresh_count -= 1
                    queue.append((nr, nc))

    return minutes if fresh_count == 0 else -1

Recommended Practice Order

1. Traversal Fundamentals: Practice spiral matrix, diagonal traversal, and search in a sorted matrix.
2. BFS/DFS Applications: Solve number of islands, rotten oranges, and shortest path in binary maze.
3. Dynamic Programming: Work on unique paths, minimum path sum, and maximal square.
4. Advanced Patterns: Tackle problems like matrix rotation, set matrix zeroes (in-place), and word search.

This progression builds from basic control flow to complex optimization, covering the core patterns Flipkart assesses.

Practice Matrix at Flipkart