Building AI Agents: From Loop to Product

By the end

What you'll build

• Explain the think-act-observe agent loop and implement it by hand, including the final-answer and step-cap brakes.
• Apply ReAct and up-front planning appropriately, and decompose goals into deterministically checkable sub-tasks.
• Design agent memory - working, episodic, and semantic - using sliding windows and summary buffers.
• Build multi-tool agents over a mock CRM, calculator, and document corpus that are graded on final environment state.
• Diagnose agent failures from traces, including corrupt success and prompt injection, and classify them precisely.
• Implement guardrails, tool allowlists, retries with backoff, and idempotency to make tool use safe.
• Read the patterns behind LangGraph, CrewAI, and the OpenAI Agents SDK without depending on any single framework.
• Write state-based evaluation assertions instead of judging an agent by its prose.
• Deploy a real agent from an official quickstart and manage keys, cost, and observability responsibly.
• Present sandbox and deployed work honestly in an agent-engineering interview, with no reliance on job or salary promises.

Curriculum

What's inside

8 modules · 44 lessons

1. 01

   The Agent Loop and ReAct

   6 lessons
   • What an Agent Really Is (and the Honest Line)
   • ReAct: Reasoning and Acting, Interleaved
   • Hand-Roll a ReAct Dispatcherpython lab
   • Your First Real Loop: A Billing Assistantagent lab
   • Check: The Loop and ReActquiz
   • + 1 more lesson
2. 02

   Planning and Decomposition

   6 lessons
   • Why One Prompt Isn't a Plan
   • Task Decomposition Patterns
   • Decompose a Goal into an Ordered Planpython lab
   • A Planning Agent Over a Mock CRMagent lab
   • Planning Vocabularyfill blank
   • + 1 more lesson
3. 03

   Memory

   6 lessons
   • Agent Memory: Working, Episodic, Semantic
   • Context Windows, Summarization, and Scratchpads
   • A Sliding-Window Summary Memorypython lab
   • An Agent That Remembers a Customeragent lab
   • Memory Types and Techniquesflashcard sr
   • + 1 more lesson
4. 04

   Guardrails, Retries and Failure Handling

   7 lessons
   • How Agents Fail: A Taxonomy
   • Error Analysis: Find the Corrupt Successtrace lab
   • Guardrails: Validation and Tool Allowlists
   • Retries, Timeouts, and Idempotency
   • Retry with Backoff and an Idempotency Guardpython lab
   • + 2 more lessons
5. 05

   Framework Pattern Literacy

   6 lessons
   • Reading Frameworks Like an Engineer: LangGraph
   • CrewAI: Roles, Tasks, and Crews
   • The OpenAI Agents SDK and the Agents Loop
   • Map the Framework Primitivesfill blank
   • Write a Role-Based System Promptprompt lab
   • + 1 more lesson
6. 06

   Capstone: Build a Support Agent

   6 lessons
   • Capstone Brief: A Support Agent for a Mock SaaS
   • Designing the Retrieval-then-Action Flow
   • Capstone: The Late-Delivery Refund Agentagent lab
   • Write the Eval Assertions for Your Agentpython lab
   • Capstone and Evaluation Conceptsflashcard sr
   • + 1 more lesson
7. 07

   Shipping Beyond the Sandbox

   4 lessons
   • From Sandbox to Real: What Actually Changes
   • Deploy a Real Agent: The Official Quickstarts
   • Production Readiness Checklistflashcard sr
   • Evidence Check: You Ran a Real Quickstartquiz
8. 08

   Agent-Engineer Interview Prep

   3 lessons
   • What Agent-Engineer Interviews Actually Test
   • Mock Interview: Agent Engineering Fundamentalsinterview lab
   • Mock Interview: Design an Agent (System Design)interview lab

The shape of it

How this course works