Remove duplicates from a table
Out loud, in order: clarify, build it, check it, explain it.
Practise SQLLoadingone moment…
Data engineering interview · Finance, Payments & Fintech
Deep Spark internals and cluster sizing, scenario SQL (self-joins, window-with-range), and classic coding problems.
01
And what its data engineers work with.
02
Stages in the order candidates meet them.
5 stages. 3 rounds have reported questions, described below.
03
The topics it covers and the questions candidates reported.
4 reported questions. See them below
central
High confidence5 reported questions. See them below
3 reported questions. See them below
04
By round. Questions marked Expected are written from the company's standards.
SQLHigh confidence
SparkHigh confidence
CodingMedium confidence
Nobody reported these exact questions at Paytm. They are the questions a loop for each of its top topics is known to ask, so prepare for them as likely, not as confirmed.
05
Ranked by how much of the evidence points at each topic.
06
Do the first group before the second.
Spark internals + cluster sizing (executors/cores/memory for 500GB), join optimization, SCD, file formats, perf debugging
Do first (3)Scenario SQL (dedup, self-joins, window-with-range, statement design)
Coding (subarray sum K, merge K sorted lists)
Hadoop/Hive/Airflow/Scala
ThenPayments pipeline design
Behavioral
07
Why candidates were turned down, as reported.
Cannot reason about Spark performance degradation or cluster sizing
Spark is named in 25 of 50 company accounts.Weak join optimization/SCD
SQL is named in 37 of 50 company accounts.Shaky scenario SQL
SQL is named in 37 of 50 company accounts.08
4 reported, 1 composite. Each opens and closes. Composites are labelled.
An online test, then technical rounds, then behavioral. SQL: remove duplicates, a self-join on cricket data, a window with a range, and design a bank statement. Spark: join tuning, slowly changing dimension types, file formats, finding why a job went from 5 minutes to 2 hours, repartition against coalesce and sizing a cluster for 500 GB. Coding: word frequency, subarray sum K, merge K sorted lists.
What to take from it Spark tuning and data structures both matter.
Four rounds: two technical, one manager, one HR. Round one covered SQL basics (joins, group by, order by, rank), questions on counting join results and simple Python such as finding anagrams in a list.
What to take from it Know how many rows each join type returns.
The second technical round was about Spark: RDDs, drivers, persistence and tuning Spark applications. The manager round was mostly situation and behavior questions.
What to take from it Be able to explain how you tune a Spark job.
The role asked for two to three years of experience. The candidate said they were chosen mostly for data structures and algorithms, plus Spark and Hadoop. They prepared DSA, Java, Spark, Hadoop, AWS and Hive.
What to take from it Prepare data structures even for a data role.
Composite walkthroughs. Built from the company's documented loop, not from one person's account.
This is a composite, not one person's account. It is built from the loop we have on record for Paytm. Questions reported for this round: Remove duplicates from a table; IPL team match combinations with conditions (self joins); Window functions with a RANGE.
What to take from it Practise these questions before you reach this round.
Strong first-hand signal; Spark tuning + cluster sizing math are the differentiators.
09
Their reported questions, one at a time, against a stopwatch.
Remove duplicates from a table
Out loud, in order: clarify, build it, check it, explain it.
Practise SQL10
Pick your level and the weeks you have.
Built from what Paytm asks, strongest first. For the wider basics at your level, follow the Mid level plan alongside it.
Spark and big data
Read: Spark chaptersDrill: Spark playgroundSay it aloud: Spark questionsFull plan: Ace the Spark round
Pipeline and system design
Read: Data pipeline chaptersDrill: Pipeline practiceSay it aloud: Data pipeline questionsFull plan: Ace the pipeline round
Data modeling
Read: Data modeling chaptersDrill: Modeling practiceSay it aloud: Data modeling questionsFull plan: Ace the modeling round
Python and coding
Read: Python chaptersDrill: Python practiceSay it aloud: Python questionsFull plan: Ace the Python round
Behavioral and communication
Read: Behavioral storytellingSay it aloud: Behavioral questions
Rehearse the Paytm loop
Run the question deckThen repeat your weakest topic once more.