The DSA Sheet: 370 Problems From Striver, Love Babbar and Apna College, Merged and Ordered by Role

Search "DSA sheet" and you get three answers that disagree with each other: Striver's A2Z sheet, Love Babbar's 450, and the Apna College sheet. Each one is good. Each one is also long, and each one assumes you are preparing for the same generic SDE interview. This post is about the sheet DSA Tracker actually uses: the three merged into one list of 370 problems across 22 topics, with duplicates removed and the order changed for the job you want.

What is in the merged sheet

We took the three sheets, matched problems by their canonical statement (not by title, since the same problem is named differently on LeetCode and GeeksforGeeks), and kept one copy of each. The counts, straight from the dataset:

• Striver A2Z: 315 of the 370 problems come from it.
• Love Babbar 450: 248.
• Apna College: 104.
• 75 problems are in all three sheets. That overlap is the closest thing to a consensus the community has, and it is a sheet of its own: Consensus 75.

The topic split is uneven on purpose, because interviews are uneven. Dynamic Programming has 51 problems and Graph has 48; Queue has 4 and Intervals has 5. A sheet that gives every topic ten problems is lying to you about where the difficulty is.

Why one sheet is not enough on its own

The three sheets were written for the same reader: a student preparing for a product-company SDE loop. If that is you, Striver's order is fine. The problem is everyone else. A frontend candidate who follows any of the three sheets top to bottom spends their first weeks on the wrong material, because the sheets front-load arrays and then march through trees and graphs regardless of what a frontend round asks.

So the merged sheet does one extra thing: every one of the 22 topics is tagged core, relevant, or optional for each of 29 roles. The problems do not change. The order does. A frontend engineer opens the sheet and sees Arrays, Strings, Two Pointers and Sliding Window first; an embedded engineer sees Bit Manipulation at the top, which a backend order puts near the bottom.

How to use it

1. Pick a role, not a sheet. The role decides the order. If you genuinely do not know yet, start with Consensus 75: 75 problems every sheet agrees on, no decision required.
2. Work topic by topic, not problem by problem. Solve six to eight problems in one topic back to back until the trigger is automatic. Two Sum, Contains Duplicate and Group Anagrams are one idea (a hash map) told three ways; doing them together teaches the idea.
3. Read our note before your first attempt, not after. Every problem page has a short intuition, the approaches from brute force to optimal with their complexity, and the pitfalls that fail hidden tests. Trapping Rain Water is the classic example of a problem where the brute force is obvious and the two-pointer version is not.
4. Revise on a schedule. A solved problem is forgotten in about two weeks if you never see it again. The tracker asks you to rate your confidence after each revision and spaces the next one out accordingly.

If you already follow one of the three sheets

You do not have to switch. The tracker keeps each sheet as a filter, with your progress on it: Striver A2Z, Love Babbar 450, Apna College. Progress on a problem counts on every sheet that contains it, so finishing Striver's arrays section also fills in the same problems on Babbar's.

What this sheet is not

It is not a shortcut. 370 problems is still 370 problems. What changes is the chance that problem number 40 is one your interview actually asks, instead of one the sheet's author happened to like. And it is not original content: the problems belong to their platforms, and every problem page credits the source. What we add is the merge, the order, the notes, and the in-browser tracer that steps through your own solution line by line.