Description
Vizuara's harness-engineering workshop studies the execution layer around a language model: the agent loop, tools, permissions, context and recovery mechanisms. Its five-session structure begins by distinguishing prompts, context engineering and the wider harness, then moves into tool management and isolated execution environments.
Later sessions address context budgets, compression, memory, session recovery and subagent coordination. Architecture reviews and a final project connect these components into a working agent-management design. The focus is on how the surrounding software controls and sustains an agent, rather than on model training alone.
Intended audience: Python or TypeScript developers familiar with basic LLM API calls. Engineers building automated tools and agent infrastructure.
What you will learn
- Explain the responsibilities of an agent harness and its main loop.
- Design tool permissions and isolated execution boundaries.
- Manage context budgets, compression and memory.
- Examine recovery, subagent coordination and complete project architecture.
Course specifications
- Publisher
- vizuara
- Instructors
- Vizuara
- Lecture count
- 5
- Quality
- 480p
- Level
- Beginner to Advanced
- Duration
- 10 hours and 23 minutes
- Subtitle
- None
- Edition
- 2026-8
Course topics
View course topics
The course topics image is temporarily unavailable.
Pictures
This picture is temporarily unavailable.
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
1.6 GB
Leave a Reply
Signed-in member comments appear immediately.