CF

AI Agents & Workflows - The Practical Guide

4h 2m 13s
English
Paid

AI agents can sound like a buzzword. But the ideas behind them are clear and useful. You can use them to change data, write content, support users, and run research tasks. When you learn how they work, you can build tools that fit your real needs.

This course shows you how to build these tools step by step. You learn how AI agents and AI workflows work, where they differ, and when to use each one. You also see how models, code, and data fit together to form a full AI system.

What You Learn

You get clear rules, simple theory, and code that you can test at once. You start with the core ideas behind agents and workflows. Then you move to real cases that show how these ideas work in practice.

  • Plain and structured guides on AI workflows and agents
  • Many practical examples like content tools, support bots, and research helpers
  • Concepts that work in any programming language

Skills You Build

You will build agents and workflows from scratch. You will use the OpenAI API in Python, and learn patterns you can reuse in other languages.

  • Create AI workflows and agent systems
  • Work with OpenAI models through API and SDK
  • Change and prepare input data with AI
  • Build AI‑based automations
  • Connect your systems to services like Slack
  • Use AI for self-check and result review
  • Manage short and long-term memory for agents
  • Build multi-agent systems and share data between them
  • Split work between general and focused agents
  • Add Human‑in‑the‑Loop steps to improve quality

All examples use Python with the OpenAI API and SDK. The ideas stay the same in other languages, so you can apply them to your own stack.

Additional

https://github.com/mschwarzmueller/ai-agents-workflows-course/tree/main

About the Author: Academind Pro (Maximilian Schwarzmüller)

Academind Pro (Maximilian Schwarzmüller) thumbnail

Academind is the teaching brand of Maximilian Schwarzmüller (Max) and Manuel Lorenz, two German developers whose Udemy catalog has become one of the largest single-instructor presences on that platform. Max in particular is widely cited as one of the clearest teachers of the JavaScript framework landscape — his Angular, React, Vue, and Node.js courses have collectively taught millions of students.

The Academind Pro platform extends beyond Udemy with deeper, more comprehensive courses aimed at developers building real applications rather than picking up syntax. Course material covers the full modern web stack: React (including Next.js), Vue, Angular, Node.js, NestJS, TypeScript, Docker, AWS, React Native, Flutter, and the broader full-stack JavaScript ecosystem.

The CourseFlix listing under this source carries over 25 Academind Pro courses spanning that range. Material is paid; Academind Pro runs on per-course pricing on the original platform. Courses are taught in Max's signature thorough, build-an-application-with-me style — long-form, deeply project-based, and continuously updated as the underlying frameworks evolve.

Watch Online 49 lessons

This is a demo lesson (10:00 remaining)

You can watch up to 10 minutes for free. Subscribe to unlock all 49 lessons in this course and access 10,000+ hours of premium content across all courses.

View Pricing
0:00
/
#1: Welcome To The Course!
All Course Lessons (49)
#Lesson TitleDurationAccess
1
Welcome To The Course! Demo
01:16
2
What Are AI Agents & Workflows?
04:56
3
An AI Agent You All Know
02:20
4
About This Course - What To Expect
02:10
5
Module Introduction
02:22
6
No Code vs With Code
02:08
7
Building AI Apps & Using AI Programmatically
04:13
8
Proprietary vs Open (Local) LLMs
05:52
9
Understanding Our Development Environment
02:30
10
Creating a New Python Project (using "uv")
01:44
11
OpenAI Setup & Pricing
05:42
12
Getting Started With A First Example Workflow
02:45
13
Preparing HTTP Requests For The OpenAI API
08:41
14
Choosing & Using a Model
02:05
15
Prompt Engineering
04:36
16
Extracting & Using the LLM Response
04:51
17
Use Those Docs!
01:39
18
Using The OpenAI Python SDK
05:50
19
Leveraging Few-Shot Prompting
04:03
20
Generating Prompts Dynamically With Dynamic Content
02:13
21
Building Multi-Step & Multi-Model Workflows
06:48
22
Workflows vs Agentic Systems
01:35
23
Using Locally Running Open Models via Ollama
08:09
24
Enforcing & Using Structured Outputs
10:54
25
Structured Outputs via SDK & Pydantic
03:57
26
Onwards To Another Example
05:39
27
Generating Images In a Workflow
06:28
28
Controlling Workflow Execution with Control Flow Adjustments
02:46
29
Control Flow In Action
08:43
30
Adding a "Human In The Loop"
06:52
31
Integrating External Services - Example: Slack
06:22
32
Module Introduction
01:37
33
How LLMs (Do Not) Use Tools
06:05
34
Implementing Tool Use From Scratch
11:25
35
Using OpenAI's Function Calling Feature
10:24
36
Building a Multi-Tool Versatile Agent
11:17
37
Building Reusable Elements With Classes
07:54
38
Getting Started with a Multi-Agent System
07:15
39
Building & Connecting Specialized Agents
10:25
40
Universal vs Specialized Agents
03:39
41
Agent Memory: Short-Term & Long-Term
05:21
42
Wrap Up
01:10
43
Module Introduction
02:37
44
Getting Started With CrewAI
04:57
45
Understanding CrewAI Agents
05:54
46
Using CrewAI Tasks
03:26
47
Adding Tools To Agents
04:15
48
Running The Crew
03:12
49
Course Roundup
01:11
Unlock unlimited learning

Get instant access to all 48 lessons in this course, plus thousands of other premium courses. One subscription, unlimited knowledge.

Learn more about subscription

Books

Read Book AI Agents & Workflows - The Practical Guide

#TitleTypeOpen
1Code Deep Dives vs Provided Code Snippets PDF
2Extracting Website Content PDF
3Important： Potential Problems & Security Risks PDF
4More On JSON Schemas & Structured Outputs PDF
5More on the OpenAI API & SDK PDF
6Using Advanced AI Models PDF
7Using Open LLMs PDF
8Using Prompt Engineering To Control Output PDF

Related courses

Frequently asked questions

What prerequisites are needed before enrolling in this course?
Before enrolling, it's beneficial to have a basic understanding of programming, particularly in Python, as the course involves using the OpenAI API and Python SDK. Familiarity with HTTP requests and model selection can also be advantageous. No prior knowledge of AI agents or workflows is required, as the course covers these topics from the ground up.
What types of projects will I build during the course?
During the course, you'll build a variety of practical projects including content tools, support bots, and research helpers. You'll create AI workflows and agent systems from scratch, work with OpenAI models, and implement automations. Additionally, you'll gain experience integrating these systems with services like Slack and managing agent memory.
Who is the target audience for this course?
The course is ideal for developers interested in integrating AI into their applications, those looking to understand AI agents and workflows, and individuals aiming to use AI for automations. It's suitable for anyone with a basic programming background who wants to explore how AI can be used to change data, write content, and support user interactions.
How does the depth of this course compare to other AI courses?
This course offers a practical approach with a focus on building real-world AI tools and understanding the integration of models, code, and data. Unlike some theoretical courses, it emphasizes hands-on experience with OpenAI API, creating workflows, and managing agent memory. It does not cover advanced AI theory or machine learning algorithms in depth.
What specific tools and platforms will I learn to use?
You will learn to use the OpenAI API and Python SDK extensively. The course includes setting up and preparing HTTP requests for the API, using proprietary and open models, and working with tools like Slack for integration. You'll also explore using locally running models via Ollama and implementing human-in-the-loop processes.
What topics are not covered in this course?
The course does not cover advanced AI model training or deep learning algorithms. It focuses more on the application of pre-trained models and workflows rather than on the development of new models from scratch. There is also no in-depth exploration of other AI platforms beyond OpenAI and CrewAI.
How much time should I allocate to complete this course?
With a total of 49 lessons, the course requires a significant time commitment. Each lesson involves practical exercises and coding tasks that may extend beyond the video runtime. It's advisable to allocate several weeks, depending on your pace, to fully understand and implement the concepts and projects presented.