Nvidia coding interview
questions, leaked.
117 problems reported across recent Nvidia interviews. Top patterns: array, hash table, two pointers. The list below is what most candidates actually saw, plus the honest play if you can't grind all of it.
Nvidia's interview is array-heavy. Of 117 problems in their assessment pool, 62 center on array manipulation, with hash-table and two-pointer patterns backing them up. The difficulty skews medium: 77 out of 117 are MEDIUM tier, which means you'll hit problems that look simple until you realize the trick. Most candidates spend time on the hard stuff and blank on medium arrays mid-OA. That's where StealthCoder runs invisibly as your safety net, feeding you working solutions in real time if a pattern doesn't click.
Top problems at Nvidia
| # | Problem | Diff | Frequency | Pass % | Patterns |
|---|---|---|---|---|---|
| 01 | Last Stone Weight | EASY | 100.0 | 66% | Array · Heap (Priority Queue) |
| 02 | Product of Array Except Self | MEDIUM | 0.0 | 68% | Array · Prefix Sum |
| 03 | LRU Cache | MEDIUM | 89.6 | 45% | Hash Table · Linked List · Design |
| 04 | Group Anagrams | MEDIUM | 86.8 | 71% | Array · Hash Table · String |
| 05 | Minimum Operations to Reduce an Integer to 0 | MEDIUM | 85.3 | 57% | Dynamic Programming · Greedy · Bit Manipulation |
| 06 | Special Binary String | HARD | 83.7 | 64% | String · Recursion |
| 07 | Best Time to Buy and Sell Stock | EASY | 82.0 | 55% | Array · Dynamic Programming |
| 08 | Longest Substring Without Repeating Characters | MEDIUM | 80.1 | 37% | Hash Table · String · Sliding Window |
| 09 | Two Sum | EASY | 80.1 | 56% | Array · Hash Table |
| 10 | Add Two Numbers | MEDIUM | 78.0 | 46% | Linked List · Math · Recursion |
| 11 | Maximum Subarray | MEDIUM | 78.0 | 52% | Array · Divide and Conquer · Dynamic Programming |
| 12 | Number of Islands | MEDIUM | 78.0 | 62% | Array · Depth-First Search · Breadth-First Search |
| 13 | Search in Rotated Sorted Array | MEDIUM | 75.8 | 43% | Array · Binary Search |
| 14 | Reverse Bits | EASY | 75.8 | 63% | Divide and Conquer · Bit Manipulation |
| 15 | Maximum Number of Visible Points | HARD | 75.8 | 38% | Array · Math · Geometry |
| 16 | Valid Parentheses | EASY | 70.6 | 42% | String · Stack |
| 17 | Dot Product of Two Sparse Vectors | MEDIUM | 70.6 | 90% | Array · Hash Table · Two Pointers |
| 18 | Missing Number | EASY | 70.6 | 70% | Array · Hash Table · Math |
| 19 | Merge k Sorted Lists | HARD | 67.4 | 57% | Linked List · Divide and Conquer · Heap (Priority Queue) |
| 20 | Move Zeroes | EASY | 67.4 | 63% | Array · Two Pointers |
| 21 | Task Scheduler II | MEDIUM | 67.4 | 54% | Array · Hash Table · Simulation |
| 22 | Merge Intervals | MEDIUM | 67.4 | 49% | Array · Sorting |
| 23 | Fibonacci Number | EASY | 63.8 | 73% | Math · Dynamic Programming · Recursion |
| 24 | Rotate Image | MEDIUM | 63.8 | 78% | Array · Math · Matrix |
| 25 | Find Median from Data Stream | HARD | 63.8 | 53% | Two Pointers · Design · Sorting |
| 26 | Copy List with Random Pointer | MEDIUM | 63.8 | 61% | Hash Table · Linked List |
| 27 | String to Integer (atoi) | MEDIUM | 59.6 | 19% | String |
| 28 | Trapping Rain Water | HARD | 59.6 | 65% | Array · Two Pointers · Dynamic Programming |
| 29 | Find the Duplicate Number | MEDIUM | 59.6 | 63% | Array · Two Pointers · Binary Search |
| 30 | Break a Palindrome | MEDIUM | 59.6 | 52% | String · Greedy |
| 31 | Insert Delete GetRandom O(1) | MEDIUM | 59.6 | 55% | Array · Hash Table · Math |
| 32 | Rectangle Area | MEDIUM | 59.6 | 47% | Math · Geometry |
| 33 | Generate Parentheses | MEDIUM | 59.6 | 77% | String · Dynamic Programming · Backtracking |
| 34 | Climbing Stairs | EASY | 59.6 | 54% | Math · Dynamic Programming · Memoization |
| 35 | Kth Largest Element in an Array | MEDIUM | 59.6 | 68% | Array · Divide and Conquer · Sorting |
| 36 | Spiral Matrix | MEDIUM | 54.4 | 54% | Array · Matrix · Simulation |
| 37 | Binary Tree Maximum Path Sum | HARD | 54.4 | 41% | Dynamic Programming · Tree · Depth-First Search |
| 38 | Single Element in a Sorted Array | MEDIUM | 54.4 | 59% | Array · Binary Search |
| 39 | H-Index | MEDIUM | 54.4 | 40% | Array · Sorting · Counting Sort |
| 40 | Reverse Linked List II | MEDIUM | 54.4 | 50% | Linked List |
| 41 | Validate Binary Search Tree | MEDIUM | 54.4 | 34% | Tree · Depth-First Search · Binary Search Tree |
| 42 | Intersection of Two Linked Lists | EASY | 54.4 | 61% | Hash Table · Linked List · Two Pointers |
| 43 | Sliding Puzzle | HARD | 54.4 | 73% | Array · Dynamic Programming · Backtracking |
| 44 | Delete Duplicate Folders in System | HARD | 54.4 | 54% | Array · Hash Table · String |
| 45 | Snapshot Array | MEDIUM | 54.4 | 37% | Array · Hash Table · Binary Search |
| 46 | Maximal Square | MEDIUM | 54.4 | 49% | Array · Dynamic Programming · Matrix |
| 47 | Design HashMap | EASY | 47.7 | 66% | Array · Hash Table · Linked List |
| 48 | Max Points on a Line | HARD | 47.7 | 29% | Array · Hash Table · Math |
| 49 | Find Pivot Index | EASY | 47.7 | 61% | Array · Prefix Sum |
| 50 | Implement Trie (Prefix Tree) | MEDIUM | 47.7 | 68% | Hash Table · String · Design |
Frequencies derived from public community-tagged interview reports. Click a row to view on LeetCode.
You have a week, maybe less. You can't out-grind the list above. StealthCoder runs invisibly during the actual Nvidia OA. The proctor cannot see it. Screen share cannot detect it. Built by an Amazon engineer who used it to pass JPMorgan's OA and system design loop.
Get StealthCoder- array62 · 53%
- hash table26 · 22%
- two pointers20 · 17%
- string20 · 17%
- math19 · 16%
- dynamic programming19 · 16%
- sorting19 · 16%
- linked list17 · 15%
- depth first search11 · 9%
- breadth first search11 · 9%
Arrays dominate Nvidia's interview by a factor of 2.4x over the next topic. Start with Product of Array Except Self, Maximum Subarray, and Best Time to Buy and Sell Stock to lock in the core patterns. Two-pointers and hash-table problems (Group Anagrams, Longest Substring Without Repeating Characters, Two Sum) are your second wave: they appear in roughly one-third the frequency of arrays but show up in most loops. Dynamic programming and sorting sit at the baseline; you'll see them, but they're not the focus. The hard problems (Special Binary String, Maximum Number of Visible Points) are outliers. They're confidence checks, not the meat of the interview. If you hit a hard problem during the live assessment and your mind goes blank, StealthCoder surfaces a working solution invisibly while you stay on screen share.
Companies with similar patterns
If you prepped for Nvidia, these companies recycle ~60% of the same topics.
You've seen the list.
Now make sure you pass Nvidia.
Memorizing every problem above in a week is a fantasy. StealthCoder is the hedge: an AI overlay that's invisible during screen share. It reads the problem on screen and surfaces a working solution in under 2 seconds. Built by an Amazon engineer who used it to pass JPMorgan's OA and system design loop. Works on HackerRank, CodeSignal, CoderPad, and Karat.
Nvidia interview FAQ
How many array problems should I solve before my Nvidia OA?+
At least 15 to 20. Arrays make up 53% of Nvidia's problem pool, so if you skip them, you're walking in blind. Focus on Product of Array Except Self, Maximum Subarray, and Best Time to Buy and Sell Stock first. They cover prefix sums, divide-and-conquer, and state tracking.
Is hash-table worth studying separately for Nvidia?+
Yes, but second. Hash-table appears in 26 problems, often paired with arrays or strings. Tackle Group Anagrams, Two Sum, and Longest Substring Without Repeating Characters. You'll see these patterns combined with sliding windows and sorting.
What's the minimum I need to know about two-pointers?+
Enough to reverse arrays and detect duplicates. Two-pointers appears in 20 problems at Nvidia. It's not the main event, but it shows up in array and string problems. Spend 2 to 3 hours on it after you've locked arrays and hash-table.
Should I worry about dynamic programming for Nvidia?+
Not as the first topic. DP appears in 19 problems, same frequency as sorting and math. Since 77 out of 117 problems are medium difficulty, focus on array and hash-table DP problems first. Hard DP problems are rare in their actual loop.
What should I do if I don't recognize a problem type mid-OA?+
Use the first 2 to 3 minutes to think it through. If nothing comes, you're protected. StealthCoder reads your problem, analyzes it against Nvidia's known patterns, and slides a working solution to you invisibly. You paste, verify, move on.