Data Generation with Pandas | Dataford Interview Questions

Using Pandas and NumPy, generate a 100,000-row dataset with columns ID (1 to 100000), a randomly selected sentence from a given list, a random cost between 0 and 100, and the sentence length; print the first 5 rows.

Asked in the Technical round 1 stage. Explain approach on paper first, then code in an online IDE. Reported follow-ups: given an LLM context window K, split the DataFrame into the minimum number of DataFrames so that total sentence length in each split does not exceed K, also considering minimum cost per split.

Output

Return the first 5 rows only.
Include the columns id, sentence, cost, and sentence_length.
Rows should be ordered by id ascending.
Ignore any row where sentence is NULL.

Schema

sentence_dataset

Column	Type	Description
idPK	INTEGER	Row identifier from 1 to 100000
sentence	VARCHAR(255)	Randomly selected sentence text
cost	NUMERIC(5,2)	Random cost between 0 and 100
sentence_length	INTEGER	Character length of the sentence

Your question is Data Generation with Pandas. Start with the requirements and the one table in the Question tab.

Run and submit as often as you like. When you're ready, talk me through your approach or go straight to the code.

Problem

Output

Return the first 5 rows only.
Include the columns id, sentence, cost, and sentence_length.
Rows should be ordered by id ascending.
Ignore any row where sentence is NULL.

Schema

sentence_dataset

Column	Type	Description
idPK	INTEGER	Row identifier from 1 to 100000
sentence	VARCHAR(255)	Randomly selected sentence text
cost	NUMERIC(5,2)	Random cost between 0 and 100
sentence_length	INTEGER	Character length of the sentence

Your question is Data Generation with Pandas. Start with the requirements and the one table in the Question tab.

Run and submit as often as you like. When you're ready, talk me through your approach or go straight to the code.