Question 1

What is the difference between a generator and a list in Python?

Accepted Answer

**List**: Materializes all elements in memory; indexable; reusable. **Generator**: Yields one element at a time; lazy; single-pass; memory O(1) for the sequence. **Why it matters**: Generators avoid loading huge datasets into memory; essential for streaming and large files. **Scalability trade-off**: Generators can't be sliced or revisited without re-creation; lists support random access and multiple passes....

Question 2

Can you explain the concept of polymorphism and inheritance in Java with examples?

Accepted Answer

**Inheritance:** Class extends another; inherits fields/methods. `class Dog extends Animal`. **Polymorphism:** Supertype reference → subtype; runtime dispatch. `Animal a = new Dog(); a.eat()` calls Dog's override. **Why:** Extensibility; Liskov Substitution. **Trade-off:** Composition often > inheritance. **Production:** Interfaces for contracts....

Question 3

Coin Change Problem - minimum number of coins required to make change

Accepted Answer

**DP:** `dp[i]` = min coins for amount i. `dp[0]=0`; for each coin c: `dp[i]=min(dp[i], 1+dp[i-c])`. O(amount·coins). **Greedy:** Only for canonical systems (US). **Why:** Classic DP; optimization. **Edge:** amount 0; no solution....

Question 4

Count occurrences of elements in a list of tuples using Spark RDDs

Accepted Answer

**RDD:** `rdd.flatMap(lambda t: t).map(lambda x: (x,1)).reduceByKey(lambda a,b: a+b)`. Or `countByValue()` for whole tuples. **Why:** `reduceByKey` for distributed count. **Scalability:** Partitioning key; avoid skew....

Question 5

Design a solution to generate unique device names from a list of IoT devices.

Accepted Answer

**Section 1 — The Context (The 'Why')**
The primary challenge for this design in Python/Coding is balancing scale, cost, and reliability. At scale, naive approaches fail: single points of failure cause cascades, schema evolution breaks consumers, and over-provisioning explodes cost. Failure modes include silent data loss from non-idempotent writes, cascading job failures from tight coupling, and operational burden from manual intervention....

Question 6

Design an algorithm to merge k sorted lists of video streaming data.

Accepted Answer

**Section 1 — The Context (The 'Why')**
The primary challenge for this design in Python/Coding is balancing scale, cost, and reliability. At scale, naive approaches fail: single points of failure cause cascades, schema evolution breaks consumers, and over-provisioning explodes cost. Failure modes include silent data loss from non-idempotent writes, cascading job failures from tight coupling, and operational burden from manual intervention....

Question 7

Difference between Stack vs Queue

Accepted Answer

**Stack:** LIFO; push/pop top. **Queue:** FIFO; enqueue rear, dequeue front. **Use:** Stack—backtracking, DFS. Queue—BFS, streaming. **Python:** list for stack; `deque` for both. **Why:** Different access patterns....

Question 8

Explain your approach to designing a scalable customer loyalty program data platform.

Accepted Answer

**Section 1 — The Context (The 'Why')**
The primary challenge for this design in Python/Coding is balancing scale, cost, and reliability. At scale, naive approaches fail: single points of failure cause cascades, schema evolution breaks consumers, and over-provisioning explodes cost. Failure modes include silent data loss from non-idempotent writes, cascading job failures from tight coupling, and operational burden from manual intervention....

Question 9

Given 1TB of a file, how to check word count?

Accepted Answer

**Distributed:** Spark: `textFile().flatMap(split).map(w=>(w,1)).reduceByKey(add)`. Or Hadoop MapReduce. **Single:** Stream chunks. **Why:** Won't fit memory. **Scalability:** Partition by file/split. **Production:** Spark for 1TB.

Question 10

How does your tech stack support scalability and analytics?

Accepted Answer

**Scalability:** Distributed compute (Spark, Flink); columnar (Parquet); partitioning; batch+stream separation; auto-scale. **Analytics:** Pre-aggregation; materialized views; cache (Redis, Alluxio). **Why:** Scale compute independently....

Hard Python Interview Questions

Companies asking these questions

All 28 Questions

More Interview Prep Guides

Get All Answers in PDF Format

Hard Python Interview Questions

Companies asking these questions

All 28 Questions

More Interview Prep Guides

Get All Answers in PDF Format