What is an AI agent in Agentic AI - Practice Questions & Exercises

Challenge - 5 Problems

🧠 Conceptual

intermediate

Understanding the role of an AI agent

Which of the following best describes what an AI agent does?

A It stores large amounts of data without interacting with anything.

B It senses its environment and takes actions to achieve goals.

C It only processes images without making decisions.

D It randomly generates outputs without input.

💡 Hint

Think about how a robot or virtual assistant works by noticing things and acting.

An AI agent observes its surroundings and acts to reach specific goals, like a self-driving car sensing roads and steering.

❓ Predict Output

intermediate

Output of a simple AI agent simulation

What will be the output of this simple AI agent code that moves in a 1D space?

Agentic AI

class SimpleAgent:
    def __init__(self):
        self.position = 0
    def sense(self, environment):
        return environment[self.position]
    def act(self, perception):
        if perception == 'obstacle':
            self.position -= 1
        else:
            self.position += 1

environment = ['clear', 'clear', 'obstacle', 'clear']
agent = SimpleAgent()
for _ in range(3):
    p = agent.sense(environment)
    agent.act(p)
print(agent.position)

A 1

B 2

C -1

D 0

💡 Hint

Track the agent's position step by step as it senses and acts.

The agent starts at 0, moves forward twice on 'clear', then encounters 'obstacle' and moves back once, ending at position 1.

❓ Model Choice

advanced

Choosing the right AI agent model for a task

You want to build an AI agent that learns to play a video game by trial and error. Which model type is best suited?

A Reinforcement Learning agent

B Supervised Learning classifier

C Unsupervised clustering model

D Rule-based expert system

💡 Hint

Think about learning from rewards and punishments over time.

Reinforcement Learning agents learn by interacting with the environment and receiving feedback, ideal for games.

❓ Metrics

advanced

Evaluating an AI agent's performance

Which metric best measures how well an AI agent achieves its goals over time?

A Silhouette score

B Mean squared error

C Precision

D Cumulative reward

💡 Hint

Consider a score that adds up all successes and failures during tasks.

Cumulative reward sums all rewards the agent gets, showing overall success in goal achievement.

🔧 Debug

expert

Debugging an AI agent's action logic

What error will this AI agent code raise when running?

Agentic AI

class Agent:
    def __init__(self):
        self.state = None
    def act(self, input):
        match input:
            case 'go':
                self.state = 'moving'
            case 'stop':
                self.state = 'stopped'
            case _:
                self.state = 'idle'
agent = Agent()
agent.act('stop')

A TypeError because 'input' is a reserved word

B AttributeError because 'state' is not defined

C SyntaxError due to missing colon after 'case "stop"'

D No error, code runs fine

💡 Hint

Check the syntax of the match-case statements carefully.

The 'case' line for 'stop' is missing a colon, causing a SyntaxError.