The objective of this workshop is simple: everyone should leave having built an agent.
Not a vague AI concept. Not a prompt with a nice wrapper. An actual agent with a clear job, a defined workflow, and a reason to exist.
Building the agent is the visible output. The deeper goal is to improve how people think about working with AI.
Through the process of designing and building an agent, participants should develop:
- Better judgment about where AI is useful and where it is not
- Better instincts for scoping AI systems around real workflows
- Better understanding of what should be handled by software versus by the model
- Better habits for working with AI as a tool for reasoning, iteration, and execution
- Better leverage from AI in their day-to-day work
This workshop is not just about producing an artifact. It is about expanding how participants use AI and how seriously they think about applying it.
By the end of the workshop, each participant should:
- Pick a narrow, meaningful problem
- Design an agent around that problem
- Build a first working version
- Learn what makes an agent useful instead of impressive
- Leave with a stronger framework for future AI projects
- Submit your work to
/agents - Present in the next company AI session
- What did you build?
- How did you go about it?
- What challenges did you encounter?
- What were your learnings
The standard is not "most ambitious."
The standard is:
- Clear scope
- Real utility
- Thoughtful boundaries
- Good judgment
- A working specialist agent