Data analysis agent pipeline in Agentic AI - Practice Problems & Coding Challenges

Challenge - 5 Problems

❓ Predict Output

intermediate

Output of a simple data filtering step in an agent pipeline

Consider an agent pipeline that filters a dataset to include only rows where the value in column 'age' is greater than 30. What is the output DataFrame after this filtering?

Agentic AI

import pandas as pd

data = pd.DataFrame({'name': ['Alice', 'Bob', 'Charlie', 'David'], 'age': [25, 35, 30, 40]})
filtered_data = data[data['age'] > 30]
print(filtered_data)

A

    name  age
0  Alice   25
2  Charlie 30

B

    name  age
1    Bob   35
3  David   40

C

    name  age
1    Bob   35
2  Charlie 30
3  David  40

D

Empty DataFrame
Columns: [name, age]
Index: []

💡 Hint

Remember the condition is strictly greater than 30, so age 30 is excluded.

The filter selects rows where 'age' > 30, so only Bob (35) and David (40) remain.

❓ data_output

intermediate

Result of aggregation in a data analysis pipeline

An agent pipeline groups data by 'department' and calculates the average salary. What is the resulting DataFrame?

Agentic AI

import pandas as pd

data = pd.DataFrame({'department': ['HR', 'IT', 'HR', 'IT', 'Finance'], 'salary': [50000, 60000, 55000, 65000, 70000]})
grouped = data.groupby('department')['salary'].mean().reset_index()
print(grouped)

A

  department   salary
0   Finance  70000.0
1        HR  52500.0
2        IT  62500.0

B

  department   salary
0   Finance  60000.0
1        HR  55000.0
2        IT  65000.0

C

  department   salary
0   Finance  70000
1        HR  105000
2        IT  125000

D

  department   salary
0   Finance  70000.0
1        HR  55000.0
2        IT  60000.0

💡 Hint

Average salary means sum of salaries divided by count per department.

HR average: (50000+55000)/2=52500, IT average: (60000+65000)/2=62500, Finance: 70000 alone.

❓ visualization

advanced

Identify the correct plot output from a data analysis agent pipeline

An agent pipeline creates a bar plot showing total sales per product category. Which option correctly describes the plot output?

Agentic AI

import pandas as pd
import matplotlib.pyplot as plt

data = pd.DataFrame({'category': ['A', 'B', 'A', 'C', 'B'], 'sales': [100, 200, 150, 300, 250]})
totals = data.groupby('category')['sales'].sum()
totals.plot(kind='bar')
plt.show()

A A bar chart with categories A, B, C on x-axis and bars showing sales totals 250, 450, 300 respectively.

B A line chart with categories A, B, C on x-axis and sales totals 100, 200, 300 respectively.

C A pie chart showing sales distribution with slices for categories A, B, C with values 100, 200, 300.

D A scatter plot with sales on x-axis and categories on y-axis.

💡 Hint

Sum sales per category: A=100+150=250, B=200+250=450, C=300.

The code groups sales by category and plots a bar chart of sums: A=250, B=450, C=300.

🔧 Debug

advanced

Identify the error in a data cleaning step of an agent pipeline

An agent pipeline tries to fill missing values in a DataFrame column 'score' with the mean score but raises an error. What is the error?

Agentic AI

import pandas as pd

data = pd.DataFrame({'score': [10, None, 30, None, 50]})
mean_score = data['score'].mean()
data['score'] = data['score'].fillna(mean_score())
print(data)

A No error, outputs filled DataFrame

B KeyError: 'score' not found

C ValueError: cannot convert float NaN to integer

D TypeError: 'float' object is not callable

💡 Hint

Check how mean_score is used in fillna.

mean_score is a float value, calling mean_score() causes TypeError.

🚀 Application

expert

Determine the final output of a multi-step data analysis agent pipeline

An agent pipeline performs these steps on a DataFrame: 1) filters rows where 'value' > 10, 2) creates a new column 'value_squared' as square of 'value', 3) groups by 'category' and sums 'value_squared'. What is the final output DataFrame?

Agentic AI

import pandas as pd

data = pd.DataFrame({'category': ['X', 'Y', 'X', 'Y', 'Z'], 'value': [5, 15, 20, 8, 25]})
filtered = data[data['value'] > 10]
filtered['value_squared'] = filtered['value'] ** 2
grouped = filtered.groupby('category')['value_squared'].sum().reset_index()
print(grouped)

A

  category  value_squared
0        X           400
1        Y           64
2        Z           625

B

  category  value_squared
0        X           425
1        Y           225
2        Z           625

C

  category  value_squared
0        X           400
1        Y           225
2        Z           625

D

  category  value_squared
0        X           400
1        Y           225

💡 Hint

Calculate squares only for values > 10, then sum per category.

Filtered rows: Y=15, X=20, Z=25; squares: 225, 400, 625; sums per category: X=400, Y=225, Z=625.