What this lesson covers
Quiz: What Is an AI Agent?
A practical, ground-up introduction to AI agents. Understand the ReAct loop, tools, orchestration, memory, and evaluation, then build real agents using Aurora Trading Agent platform as your lab.
Explain the difference between a bare language model, a chatbot, and an AI agent
Learning outcome for this course.
Describe the ReAct loop (Thought → Action → Observation) and why it makes agents agentic
Learning outcome for this course.
Design tool-calling workflows and understand why tools, function calling, and MCP servers are all the same core concept
Learning outcome for this course.
Use autonomy controls (whitelists, approvals) to safely run agents in production
Learning outcome for this course.
Connect memory, scheduling, and subagents into a full autonomous workflow
Learning outcome for this course.
Build traces and evaluators (including LLM judges) to measure agent quality
Learning outcome for this course.
Practice questions
Answer inside the lesson player. Correct-answer keys and explanations are not exposed in this public preview.
What is the main difference between a bare large language model (like the OpenAI Playground) and ChatGPT?
ChatGPT uses a more advanced model · ChatGPT wraps the model with tools like web search and code execution · The OpenAI Playground can't generate text · There is no difference. They're the same thing
A language model is stateless. What does this mean?
It can't generate structured output like JSON · It doesn't remember anything between calls. Every interaction starts from zero · It can only respond in English · It requires an internet connection to work
What is a 'tool' in the context of AI agents?
A physical device the AI controls · A function, API call, or command that the AI can trigger to get information or take action · A special type of language model · A user interface for chatting with AI
What makes an AI agent different from a chatbot with tools?
Agents use more expensive models · Agents operate in a loop: they observe results, reason about them, act, and repeat until the task is complete · Agents can only work with JSON data · Agents don't need a language model
In prompt engineering, what is the difference between zero-shot and few-shot prompting?
Zero-shot is free and few-shot costs money · Zero-shot gives the model only instructions. Few-shot also includes examples of correct input/output pairs so the model can match the pattern · Zero-shot uses one model and few-shot uses multiple models · There is no difference. They are the same technique
The system prompt is the primary way you give a language model its context and control its behavior.
True · False
When an AI agent generates a JSON tool call, the JSON itself performs the action.
True · False