LeetCode mapped to patterns
A curated list maps 96 LeetCode problems to 16 common algorithmic patterns to help focused practice rather than random grinding. The collection highlights families like Sliding Window, Two Pointers, Binary Search, Hashing, Heaps and Dynamic Programming and offers a full PDF in the thread. (x.com)
A growing interview-prep playbook is telling candidates to stop solving coding problems at random and start by pattern. (github.com) Sean Prashad’s LeetCode Patterns groups problems by recurring templates such as arrays, hash tables, linked lists, trees, depth-first search, breadth-first search, binary search and dynamic programming. The project’s README says the goal is repeated practice on common patterns instead of “randomly tackling questions.” (github.com) The public site listed 178 questions on April 13, 2026, with filters for difficulty, pattern, company and review status. The same page showed interview-frequency tags pulled from LeetCode Premium, including counts for companies such as Google, Amazon, Meta and Microsoft over the prior six months. (seanprashad.com) The idea behind pattern-based practice is simple: many coding interview prompts are variations of the same move. A “sliding window” problem expands and shrinks a range in an array or string, while “two pointers” walks from two positions to cut brute-force comparisons. (github.com) That framing has spread well beyond one repository. LeetCode now has its own pattern-based problem list, and multiple public cheat sheets and forks organize questions the same way for interview study. (leetcode.com) (github.com 1) (github.com 2) The GitHub repository shows 11,800 stars and 1,900 forks, a rough measure of how widely the checklist has circulated among candidates and coaches. The project has also been updated recently, with GitHub showing changes to question metadata and site code in April 2026. (github.com) Prashad’s README says the list was “heavily inspired” by Grokking the Coding Interview, the Blind 75 list and a HackerNoon article on interview patterns. That places the project inside a broader shift from giant question banks toward shorter, repeatable sets built around recognition. (github.com) The appeal is not that 16 or 20 patterns solve every problem exactly once. It is that candidates can learn to spot a small number of reusable structures before they memorize hundreds of answers. (github.com)
