Promise Pool
A medium-tier problem at 80% community acceptance, tagged with . Reported in interviews at Uber and 1 others.
Promise Pool shows up in Uber and Yandex assessments and catches people off guard because it looks like async control flow but demands you think like a resource manager. You're not just chaining promises, you're limiting concurrent execution so the system doesn't choke on thousands of parallel requests. The acceptance rate sits at 80%, which sounds high until you realize most candidates either spawn unlimited promises or serialize everything and hit timeouts. StealthCoder is the invisible edge if you blank on the pool pattern during your live OA and need a working implementation in seconds.
Companies that ask "Promise Pool"
Promise Pool 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 because the OA filter rejects engineers who'd pass the on-site. That's a broken filter. This is the workaround.
Get StealthCoderThe core trick is understanding that you can't just await all promises at once. You need a queue and a worker pool that processes N tasks in parallel, pulling from the queue as workers free up. Most candidates either miss the semaphore logic or implement a correct solution that's so verbose they run out of time. The pattern usually involves maintaining an array of in-flight promises and a pending task queue, popping new work onto the in-flight array when a slot opens up, then waiting for all to resolve. The gotcha is handling rejection and ensuring the pool drains completely, even if some tasks fail. Since Uber deals with concurrent request handling at scale, this problem tests whether you understand real backend bottlenecks. StealthCoder surfaces a clean, tested pool implementation the instant you need it.
You know the problem.
Make sure you actually pass it.
Promise Pool 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 because the OA filter rejects engineers who'd pass the on-site. That's a broken filter. This is the workaround. Works on HackerRank, CodeSignal, CoderPad, and Karat.
Promise Pool interview FAQ
Is Promise Pool just async/await practice or does it test something deeper?+
It tests concurrency control, not async syntax. You could write async/await all day and still serialize work or blow out memory. Uber and Yandex ask it because managing resource pools is a real backend problem. The acceptance rate of 80% masks the gap between 'it runs' and 'it handles edge cases under load.'
What's the most common way candidates fail this?+
Either Promise.all on every task at once (memory explosion and defeats the purpose) or processing sequentially (defeats parallelism). The middle ground, maintain a fixed-size in-flight set and feed the queue as workers complete, is not obvious without seeing the pattern once.
Do I need to worry about rejection and error handling here?+
Yes. A real pool must not die if one task fails. You typically catch and store errors, let the pool keep running, and either return them with results or fail the entire pool depending on the spec. The problem statement matters here.
How does Promise Pool relate to rate limiting or backpressure?+
It's the same problem from a different angle. A pool limits how many requests hit a downstream service at once. If you're integrating with an API that throttles you, a promise pool is how you stay under the limit while keeping throughput high.
Will this appear in my Uber or Yandex OA and should I memorize the solution?+
It appears in their assessments. Memorizing the exact code won't help much because the constraints or task signature vary. Understand the queue-plus-worker-pool mental model so you can rebuild it from first principles fast.
Want the actual problem statement? View "Promise Pool" on LeetCode →