sautrikroy17/dsa-interview-playbook

This repository is a curated study guide for people preparing for coding interviews at large tech companies like Microsoft, Amazon, Google, Meta, and Apple. Rather than listing hundreds of random problems, it focuses on a smaller set of algorithmic patterns that these companies actually test. The author calls it an 80/20 approach, meaning a targeted fraction of patterns covers most of what comes up in real interviews. The guide is structured into three steps. The first step points readers to Microsoft's free developer training hub to shore up core logic and data structure fundamentals before tackling interview problems. The second step covers ten algorithmic patterns that appear most often in technical interviews: sliding window, two pointers, fast and slow pointers, merge intervals, modified binary search, tree and graph breadth-first search, tree and graph depth-first search, subsets and backtracking, heap-based top-K problems, and dynamic programming. The third step, labeled advanced structures, is folded into the ten patterns rather than being a separate section. For each pattern, the guide lists specific problems from LeetCode with their difficulty level (Easy, Medium, or Hard). For example, under sliding window it lists seven problems ranging from finding the longest substring without repeating characters to the harder sliding window maximum. Under dynamic programming it covers classics like Coin Change, Longest Increasing Subsequence, and Word Break. The repository contains no code files or solutions. It is a reading list and roadmap, not a library. All linked problems point to LeetCode's platform. A few problems in the merge intervals and meeting rooms sections are marked as LeetCode Premium, meaning they require a paid subscription to access. The intended audience is anyone preparing for a software engineering technical interview who wants a focused starting point rather than an exhaustive problem set. The guide does not explain what each pattern is or how to apply it, it assumes the reader will work through the linked problems or consult additional resources to build that understanding.