AI agent training
Build your first useful AI agent.
An AI agent is an assistant that follows your instructions and, where it has been set up to, uses information and tools to help complete a task. The useful part isn’t the technology—it’s choosing the right job for it and checking what it produces.
Not every task needs a full agent.
Part of the training is knowing which level of help fits the work. Each step up offers more, and asks more of the setup.
A saved prompt
Clear, reusable instructions you can run again whenever the task comes round.
A reusable assistant
An assistant set up with instructions and reference documents for one kind of work.
A tool-using agent
Where configured, an assistant that can also use connected information and tools to help complete steps.
What the training covers.
The scope is shaped around the tools you have and the work you want to improve. Together you’ll look at:
- Choosing a suitable task
- Defining a useful output
- Writing instructions and giving context
- Working with reference documents
- Understanding tools and permissions
- Building reusable workflows
- Checking accuracy
- Testing with real examples
- Deciding where human approval is needed
Built around your tools and needs.
There’s no fixed syllabus to sit through. Tell Mark what you work with and what you’d like to achieve, and the training can be scoped from there.
Which task would you like to start with?
Share a little about your work and the tools you use.
Discuss your training needs