Question 1

How do you handle disagreements within a team?

Accepted Answer

Situation: Disagreements often stem from misaligned success criteria or incomplete information. Task: Turn subjective debate into objective decision-making. Action: I restate each position and surface underlying goals. I propose measurable criteria (latency, cost, maintainability) and gather evidence—benchmarks, POCs, incident data. For technical choices, I suggest a time-boxed spike or A/B pilot....

Question 2

Why are you looking to transition to this role/company?

Accepted Answer

Situation: Transition rationale. Task: Alignment narrative. Action: Role and company align with goals. [Company] strengths; contribute/grow. Excited about technical, team, mission. Fit between experience and needs. Intentional next step....

Question 3

Discuss API error handling and retry mechanisms.

Accepted Answer

**Why Robust Handling**: APIs fail—transient (5xx, timeout) vs. permanent (4xx). Retrying 400 wastes time; not retrying 503 causes false failures.

**Strategy**: Exponential backoff with jitter to avoid thundering herd; cap retries (3-5); circuit breaker after consecutive failures. Idempotency keys for POST/PUT.

**Cost**: Retries add latency and load; distinguish retriable vs. non-retriable. Log request IDs for correlation....

Question 4

Given a dataset, perform transformations: Filter rows where sales > 1000, Add a new column calculating a 10% discount on sales, Group data by region and calculate total revenue.

Accepted Answer

SELECT region, SUM(sales * 0.9) AS total_revenue FROM sales WHERE sales > 1000 GROUP BY region. **Filter first** to reduce rows before aggregate. **Revenue** = post-discount (sales * 0.9). **PySpark**: df.filter(col('sales')>1000).withColumn('discount', col('sales')*0.1).groupBy('region').agg(sum(col('sales')*0.9).alias('revenue'))....

Question 5

How would you fetch data from an API and load it into a DataFrame?

Accepted Answer

**Why it matters**: At scale, design choices directly impact reliability, latency, and cost. Wrong decisions compound across jobs and teams.

Fetch API → DataFrame: Use `requests` or `urllib` to call API, parse JSON, create RDD/DataFrame. Example: `import requests; r = requests.get(url); data = r.json(); df = spark.createDataFrame(data['results'])`. For pagination: loop with `next_page`; collect pages; concatenate. For large APIs: use async, batch, or streaming....

Warner Bros Discovery Data Engineer Interview Questions

+3 More Questions with Expert Answers

Other Companies

Warner Bros Discovery Data Engineer Interview Questions

+3 More Questions with Expert Answers

Other Companies