Description
Build an AI agent from its underlying parts to understand how model decisions, tools and application state interact. The video edition starts with a ReAct-style agent and develops it through retrieval, memory, planning and reflection, using a small framework called scratch_agents.
Later material addresses code execution, agent coordination and workflow patterns such as agents acting as tools or handing work to one another. Evaluation and production monitoring connect those capabilities to debugging. Context engineering is a recurring theme: deciding which information the model needs for a particular step and examining failures when that context is inadequate.
What you will learn
- Implement a basic ReAct agent and tool use.
- Add retrieval, memory, planning and reflection.
- Study multi-agent workflows and coordination patterns.
- Evaluate agent behaviour and investigate failures.
Course specifications
- Publisher
- Oreilly
- Instructors
- Younghee Song
- Language
- English
- Level
- All Levels
- Duration
- 8 hours and 31 minutes
- Subtitle
- English
- Quality
- 1080p
- Edition
- 2026-7
Course topics
View course topics
The course topics image is temporarily unavailable.
Prerequisites
- Basic Python knowledge is the stated background for the intended audience.
Pictures
This picture is temporarily unavailable.
Introduction video
How to open the files
Keep all parts of the same edition together before extracting the first archive. Open the included course media with a compatible player.
Download resources
Choose one edition and keep its parts together. Alternate providers are available when listed.
Size
1.7 GB
Leave a Reply
Signed-in member comments appear immediately.