Number of Orders in the Backlog
A medium-tier problem at 52% community acceptance, tagged with Array, Heap (Priority Queue), Simulation. Reported in interviews at Robinhood and 2 others.
Number of Orders in the Backlog is a medium-difficulty heap and simulation problem that's shown up at Robinhood, Jane Street, and Coinbase. You're managing a queue of buy and sell orders with varying prices and quantities, and you need to figure out how many orders can't be fulfilled. It looks like a simple order-matching problem until you realize the priority order matters, then the naive greedy approach falls apart. This is the kind of problem where half the people blank, then panic. If it hits your live OA cold, StealthCoder solves it in seconds invisible to the proctor.
Companies that ask "Number of Orders in the Backlog"
Number of Orders in the Backlog is the kind of problem that decides whether you pass. StealthCoder reads the problem on screen and surfaces a working solution in under 2 seconds. Invisible to screen share. The proctor sees nothing. Built by an Amazon engineer who realized the OA tests how well you memorized 200 problems, not how well you code.
Get StealthCoderThe trick is that you can't just process orders in arrival order. You need two heaps: one for buy orders (max-heap on price) and one for sell orders (min-heap on price). You match them when the max buy price >= the min sell price, reducing quantities and removing fulfilled orders. The simulation happens in real time as you iterate through the order list. Most people either miss the heap structure entirely and try a nested loop (which times out), or they don't realize orders only match when prices overlap. Common failure: trying to match on just the order itself rather than continuously checking if buy and sell heaps can be paired. Once you see the two-heap pattern, it's straightforward. The acceptance rate sits just above 50%, suggesting a lot of people get the concept but blow the implementation or edge cases around quantity tracking.
Pattern tags
You know the problem.
Make sure you actually pass it.
Number of Orders in the Backlog recycles across companies for a reason. It's medium-tier, and most candidates blank under the timer. StealthCoder is the hedge: an AI overlay invisible during screen share. It reads the problem and surfaces a working solution in under 2 seconds. Built by an Amazon engineer who realized the OA tests how well you memorized 200 problems, not how well you code. Works on HackerRank, CodeSignal, CoderPad, and Karat.
Number of Orders in the Backlog interview FAQ
Is this problem actually asked at FAANG or just fintech?+
Only fintech companies in the input data: Robinhood, Jane Street, Coinbase. No FAANG hit yet. That said, the pattern (priority queue matching across two streams) appears in real trading systems, so fintech prep files take it seriously. If you're interviewing at a trading firm, drill this.
What's the actual trick I'm missing if the brute force doesn't work?+
You need two heaps, not one. Buy orders go in a max-heap (by price), sell orders in a min-heap (by price). Loop through orders, add to the right heap, then repeatedly match the tops if buy_max >= sell_min. Reduce quantities and pop when orders are empty. The simulation runs inline, not as a separate pass.
Why is acceptance only 50% if it's just a heap problem?+
Heap syntax is easy. The hard part is realizing you need two heaps and continuous matching logic. People also mess up quantity tracking: forgetting to decrement both sides, not popping when an order is fully matched, or matching too eagerly. Edge cases bite.
How does this relate to the other heap topics I should know?+
Heap problems usually ask you to find the Kth largest or sort by one dimension. This one forces you to maintain two heaps simultaneously and match across them. It's closer to a merge or two-pointer pattern, but with priority queues. Pushes you beyond the standard heap template.
Do I need to memorize this or is it one you can figure out live?+
If you've seen two-heap matching before, you get it in minutes. If not, you'll spend 15 to 20 minutes reverse-engineering the pattern under time pressure. Since only fintech asks it, if that's your target, drill it once and own it. Otherwise, recognize the pattern and move on if you blank.
Want the actual problem statement? View "Number of Orders in the Backlog" on LeetCode →