Question 1

Implement a Double-Ended Queue (Deque) From Scratch

Accepted Answer

**Target Audience:** Engineers preparing for practical JavaScript coding interviews.
**Difficulty:** Easy

> **How to read this doc:** the ~22x speedup figure below is a real, direct measurement on a real 20,000-element deque, not a theoretical Big-O claim taken on faith.

## 1. The problem, restated

Implement a data structure supporting `pushFront`, `pushBack`, `popFront`, and `popBack` — all genuinely O(1), not just "fast enough in practice."

## 2. Clarifying questions a strong candidate ask

Question 2

Build a Toast/Notification Queue That Shows One at a Time With Auto-Dismiss

Accepted Answer

**Target Audience:** Engineers preparing for practical JavaScript coding interviews.
**Difficulty:** Easy

> **How to read this doc:** the "never more than one active" claim was verified live, in a real browser, by sampling the actual DOM at multiple points in time — not asserted from reading the code alone.

## 1. The problem, restated

Build a queue where `enqueue(message)` can be called any number of times in quick succession, but the UI only ever shows ONE toast at a time — each auto-dismiss

Question 3

Priority queue binary heap

Accepted Answer

**Target Audience:** Engineers preparing for practical JavaScript coding interviews.
**Difficulty:** Medium

> **How to read this doc:** the real, defining correctness property of a heap — that repeated `extractMin()` calls always return values in genuinely sorted order, regardless of the original INSERTION order — was verified directly with a real, randomly-shuffled 9-number sequence.

## 1. The problem, restated

A binary min-heap is a complete binary tree (stored efficiently as a plain array,

Question 4

Build an Async Priority Task Scheduler

Accepted Answer

**Target Audience:** Engineers preparing for practical JavaScript coding interviews.
**Difficulty:** Medium

> **How to read this doc:** the real, defining behavior a priority scheduler must get right — that tasks NOT YET STARTED run in priority order, while an ALREADY-RUNNING task is never interrupted — was verified directly with a real, logged execution sequence: a low-priority task allowed to keep running, followed correctly by high-before-medium for the two tasks queued after it.

## 1. The

Question 5

Implement a Promise Queue (TaskQueue) with concurrency

Accepted Answer

**Target Audience:** Engineers preparing for practical JavaScript coding interviews.
**Difficulty:** Medium

> **How to read this doc:** the "genuinely inspectable state" claim — the real, defining difference from this bank's own `p-limit` question — was verified directly by reading `running`/`pending` immediately after enqueueing, not just asserted as a design feature.

## 1. The problem, restated

Build a `TaskQueue` class: `new TaskQueue(concurrency)`, an `.add(task)` method returning a promi

Question 6

Implement p-limit — concurrency limiter for async tasks

Accepted Answer

**Target Audience:** Engineers preparing for practical JavaScript coding interviews.
**Difficulty:** Medium

> **How to read this doc:** the "never more than N active at once" claim was verified directly with a real, live execution log across 5 real async tasks at a limit of 2, not just reasoned about.

## 1. The problem, restated

`pLimit(concurrency)` returns a function; wrapping any async task-returning function with it guarantees at most `concurrency` tasks run SIMULTANEOUSLY — additional ta

Question 7

Queue via two stacks

Accepted Answer

**Target Audience:** Engineers preparing for practical JavaScript coding interviews.
**Difficulty:** Hard

> **How to read this doc:** the two-stack queue was checked against a plain array queue on 2,000 random operation sequences (0 mismatches), and the amortized claim was counted, not asserted: over 200,000 enqueues and dequeues the total number of element moves was exactly 200,000. A practical surprise was measured too: draining an array with `shift()` had a performance cliff on this Node (5.

Question 8

Throttled API poll client

Accepted Answer

**Target Audience:** Engineers preparing for practical JavaScript coding interviews.
**Difficulty:** Hard

> **How to read this doc:** the entire real point of this question — that a slow real fetch must not allow OVERLAPPING requests to fire on subsequent poll ticks — was verified directly: with a deliberately slow first fetch spanning multiple poll intervals, 2 real ticks were explicitly logged as SKIPPED rather than triggering redundant concurrent requests.

## 1. The problem, restated

`setI

JAVASCRIPT CODINGDECODED.

Easy

Implement a Double-Ended Queue (Deque) From Scratch

Build a Toast/Notification Queue That Shows One at a Time With Auto-Dismiss

Medium

Priority queue binary heap

Build an Async Priority Task Scheduler

Implement a Promise Queue (TaskQueue) with concurrency

Implement p-limit — concurrency limiter for async tasks

Hard

Queue via two stacks

Throttled API poll client