(Online Delivery) Agentic AI Builder™ (Exam AGB-110): Setting Up a Functional AI Agent

Overview:

Your organization has assessed its need and ability to implement one or more AI agents with minimal risk. Now it's just a matter of building those agents—that's where you come in. In this course, you'll translate business requirements into a functional AI agent that can automate complex tasks and processes that would otherwise require significant human effort. Ultimately, this can lead to improved user productivity, a reduction in operational costs, and enhanced employee and customer satisfaction.

This is the second course in a series and is meant to build upon the foundation of the first course by giving technical practitioners the skills they need to successfully build agentic AI into their organizations.

Course Objectives:

In this course, you will build an AI agent from scratch that can automate important tasks requiring some level of human-like judgment.

• Set up the agent development environment.
• Assess LLM behavior in agent contexts.
• Implement the agent reasoning loop.
• Implement tools for an agent to use.
• Add knowledge to an agent through retrieval-augmented generation (RAG).
• Enforce structure, safety, and reliability in an agent.
• Test the behavior and performance of an agent.
• Deploy a single-agent system to production.

Target Student:

This course is designed for programmers, IT infrastructure personnel, DevOps/MLOps personnel, and any other technically minded professional who is responsible or may soon be responsible for implementing agentic solutions in their organization.

This course is also designed to assist students in preparing for the CertNexus® Agentic AI Builder™ (Exam AGB-110) credential.

Prerequisites:

To ensure your success in this course, you should have foundational knowledge of agentic AI, including concepts like agentic frameworks, large language models (LLMs), tokens, context windows, agentic tools, retrieval-augmented generation (RAG), behavioral guardrails, and more. You can obtain this level of knowledge by taking the CertNexus AgenticAIBIZ™ (Exam AGZ-110): Foundations of Agentic AI course.

You must also have proficiency using at least one programming language. This course primarily uses Python® to illustrate agentic concepts, so experience with Python is recommended.

Course Content

Lesson 1: Setting Up the Agent Development Environment

Topic A: Configure a Python-Based Agent Workspace

Topic B: Configure LLM Access

Topic C: Configure Runtime Constraints

Lesson 2: Assessing LLM Behavior in Agent Contexts

Topic A: Analyze LLM Capabilities and Limitations

Topic B: Design Prompts for Agent Reasoning

Lesson 3: Implementing the Agent Reasoning Loop

Topic A: Implement the ReAct Pattern

Topic B: Manage Agent State Across Iterations

Topic C: Manage Memory and Persistence

Lesson 4: Using Agent Tools

Topic A: Design Agent Tools and Interfaces

Topic B: Execute and Validate Tool Calls

Lesson 5: Adding Knowledge with Retrieval-Augmented Generation

Topic A: Implement Document Ingestion and Embeddings

Topic B: Retrieve and Use Context Effectively

Lesson 6: Enforcing Structure, Safety, and Reliability in an Agent

Topic A: Enforce Structured Outputs

Topic B: Handle Uncertainty and Failures

Lesson 7: Testing an Agent

Topic A: Monitor Agent Behavior

Topic B: Evaluate Agent Performance

Topic C: Analyze an Agent for Security Flaws

Lesson 8: Deploying a Single-Agent System

Topic A: Expose Agent Interfaces

Topic B: Review Readiness and Limitations