OKX coding interview
questions, leaked.
3 problems reported across recent OKX interviews. Top patterns: array, matrix, depth first search. The list below is what most reported candidates actually saw, plus the honest play if you can't grind all of it.
OKX coding assessments are small but punishing. Three problems, two medium and one hard, all built around arrays and matrices. You'll face island-counting, matrix searches, and subarray logic under time pressure. Most candidates freeze on the hard problem because they haven't drilled monotonic queues or sliding window edge cases. StealthCoder sits invisible during the live assessment and solves whatever pattern you blank on, but you still need to walk in confident on arrays and DFS basics.
Top problems at OKX
| # | Problem | Diff | Frequency | Pass % | Patterns |
|---|---|---|---|---|---|
| 01 | Number of Islands | MEDIUM | 100.0 | 62% | Array · Depth-First Search · Breadth-First Search |
| 02 | Kth Smallest Element in a Sorted Matrix | MEDIUM | 89.1 | 64% | Array · Binary Search · Sorting |
| 03 | Count Subarrays With Fixed Bounds | HARD | 89.1 | 69% | Array · Queue · Sliding Window |
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 OKX OA. The proctor cannot see it. Screen share cannot detect it. Built by an engineer who got tired of watching his cohort grind for six months and still get filtered at the OA stage.
Get StealthCoder- array3 · 100%
- matrix2 · 67%
- depth first search1 · 33%
- breadth first search1 · 33%
- union find1 · 33%
- binary search1 · 33%
- sorting1 · 33%
- heap priority queue1 · 33%
- queue1 · 33%
- sliding window1 · 33%
Every single problem at OKX touches arrays. Two problems require matrix thinking, and the medium-difficulty tier is where most people crater because the second medium problem chains binary search and heap operations together. The hard problem, Count Subarrays With Fixed Bounds, is brutal because it demands both sliding window and monotonic queue knowledge in one shot. Your prep order is clear: master array fundamentals and matrix traversal first, then spend real time on sliding window patterns and monotonic queues before the OA. DFS and BFS are tested but secondary. If you hit the hard problem and your monotonic queue logic fragments, StealthCoder delivers a working solution in seconds, screen-share invisible.
Companies with similar patterns
If you prepped for OKX, these companies recycle ~60% of the same topics.
You've seen the list.
Now make sure you pass OKX.
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 engineer who got tired of watching his cohort grind for six months and still get filtered at the OA stage. Works on HackerRank, CodeSignal, CoderPad, and Karat.
OKX interview FAQ
Should I study DFS and BFS before arrays for OKX?+
No. Arrays appear in all three problems. Master array indexing, subarrays, and bounds-checking first. DFS and BFS matter for one medium problem, Number of Islands, but they're called out alongside array and matrix tags. Array skills are your foundation.
Is the Kth Smallest problem really binary search or is it heap-only?+
Both. The problem tags binary search, heap, and matrix, meaning the optimal solution likely combines them. Don't solve it one way. You need to understand the heap approach and the binary search approach so you can pick the right one under pressure.
How much time should I spend on monotonic queues before the OKX assessment?+
Significant time. One problem is HARD and specifically tests monotonic queues with sliding window. That's a pattern most candidates skip in generic prep. Drill 5 to 10 monotonic queue problems before you take the OA.
What's the difficulty jump from medium to the hard problem at OKX?+
Sharp. The hard problem layers sliding window constraints and monotonic queue state tracking together. The medium problems are standard matrix and array searches. If you've only drilled easy-to-medium, the hard will feel like a different language. Prepare for it specifically.
Is union-find required for OKX, or can I solve Number of Islands with just DFS?+
DFS works and is faster to code. Union-find is listed as a topic tag, so the problem supports both approaches. If you're comfortable with DFS, use it. Union-find is the hedge if your DFS recursion depth handling is weak.