Build AI Agents That Actually Work AI agents are everywhere—but most of them fail in frustrating, unpredictable ways. They get confused, waste tokens, hit dead ends, and require constant babysitting. This course teaches you the patterns and architectures that separate agents that struggle from agents that succeed. Using Python and the Model Context Protocol (MCP), you'll learn to build AI agents from the ground up—and more importantly, you'll learn why certain designs work while others fall apart. What You'll Learn: - Build MCP Tool Servers — Create the bridge that lets AI agents interact with any system: filesystems, databases, APIs, or your own custom tools - Build a Research Agent - Create your own research agent that can explore your computer and help answer questions with documents, find that missing data, or explain how code works. - Master the Agent Loop — Understand the universal pattern every AI agent follows: PERCEIVE → DECIDE → ACT → OBSERVE → REPEAT - Connect agents to tools — Wire up LLMs to discover, select, and execute tools autonomously The Patterns That Make Agents Reliable: - Response-as-Instruction — Your tools don't just return data—they guide agent behavior in real-time. Learn to design tool responses that teach the agent what to do next, when to stop, and how to communicate results. - Failing Forward — Turn errors from dead ends into stepping stones. Design error messages that teach agents how to recover—automatically, without human intervention. For the first time in computing history, your error messages have a reader that can actually do something about them. - Intelligence Budget — Every token in the context window is precious attention. Learn to maximize signal and minimize noise—pre-digesting data in tools, using scripted orchestration for mechanical work, and reserving the agent's cognitive resources for decisions that actually require intelligence.
Showing: 6 of 6
Thank you so much! This course was extremely helpful. I was able to solve a real performance/delay issue at work that was directly related to the course topic. It also helped me understand the underlying concepts much better and gave me confidence that we are approaching our implementation correctly. Highly recommended!
Polecam
a;vh
I've done a few Jules White courses and I must say that this is poor and quickly thrown together course. There are a few lectures, but most of the course is just AI generated reading material. The final exam is an AI generated test too. Overall it's a poor course. I was expecting a lot more.
The structure of the course is quite poor, and there is no clear learning path to follow. In the first module, the instructor mostly just talks without providing any slides or other visual materials to help with understanding. The rest of the modules consists largely of a lot of reading materials and code, but they are not well integrated. The code is simply given to you without being properly explained alongside the relevant materials, and there are no meaningful exercises or hands-on practice. Overall, the learning experience was not very good and felt like a waste of time. I’ve decided to switch to learning through the Anthropic Academy instead.
This course becomes extremely light on detail as soon as you get into the practical part. When you have to copy the course code, it tells you to google how to do it. And then it tells you to do something vague with the python folder...only, there is no python folder. Essentially it leads you into a dead end. I'll be checking out a different course to learn these concepts.