Top DSA interview questions & topic map

Most candidates study DSA randomly: random problems, random tags, random YouTube videos. Interview loops are not random — they sample from a skewed distribution of topics. The first chart below is an illustrative composite (typical product-company and strong campus OAs): it shows relative how often each theme appears, not a guarantee for one company on one day.

How to read the topic landscape

Arrays & hashing sit on top because they are the default substrate for OAs and many round-1 problems. Trees and graphs spike at product companies and strong intern pipelines. DP is asked less often by volume but is a high-variance filter: when it shows up, unprepared candidates collapse quickly.

Arrays, strings & hashing

Canonical patterns: frequency maps, prefix sums, sorting as preprocessing, index mapping.

• Two sum / k-sum family (hash map or sort + two pointers).
• Longest substring without repeat / at most k distinct (sliding window).
• Subarray sum equals k (prefix sum + hash map).
• Merge intervals and meeting-room style scheduling.

Two pointers & sliding window

Often combined with arrays or strings. Interviewers use these to test whether you can maintain invariants while moving indices — not whether you memorized a template.

• Container with most water, trapping rain water (two pointers from ends).
• Minimum window substring / smallest subarray with sum ≥ target.
• Fast/slow pointer for cycle detection (also linked lists).

Linked lists

• Reverse linked list (iterative + recursive).
• Merge two sorted lists, add two numbers represented as lists.
• Detect cycle (Floyd), find intersection of two lists.

Trees & BST

• Traversals, height/depth, balanced tree check, diameter of binary tree.
• LCA in BST vs LCA in binary tree.
• Path sum variants, serialize/deserialize tree.

Graphs

• Number of islands, rotten oranges, word ladder (BFS layering).
• Course schedule / topological sort.
• Connected components, DFS on grid vs adjacency list.

Dynamic programming

Start with 1D DP (climbing stairs, house robber, coin change) before 2D (LCS, edit distance). Interviewers often accept O(n²) with clear state definition over a rushed “optimized” solution you cannot explain.

Binary search & heaps

• Binary search on answer space (Koko eating bananas, split array largest sum).
• Median from data stream, merge k sorted lists, top k frequent elements.

Prep timeline: three tracks in parallel

The line chart below is illustrative: it shows how readiness might grow if you keep core DS, pattern depth, and mock interviews moving together. Mocks that start too late produce a flat green curve in real life — don’t wait until week 8 to speak out loud.

Mistakes that cost offers

1. Only coding in silence. Interviews grade communication. Narrate assumptions, complexity, and trade-offs.
2. Skipping complexity. Always state time and space; invite follow-ups on optimization.
3. One-shot practice. Re-solve the same problem after 5–7 days (spaced repetition).
4. Ignoring OAs. Many Indian campus pipelines filter hard on timed OAs — practice under timer and partial scoring mindset.

Generate role-shaped question sets with our free interview question generator, tighten behavioral stories with the STAR answer builder, and line up full sessions with InterviewEra mock interviews.