For teams, builders and professionals
Agentic AI Course: Build and Deploy AI Agents
An agentic AI course teaches you or your team to plan, build and safely run AI agents: software that uses AI to complete multi-step tasks with tools, not just answer questions.
- Format
- Live build course
- Duration
- Two days
- Location
- Sydney · Melbourne · Online
- Price
- from $7,500 AUD
- For
- Groups of 4–12, or 1:1
Outcomes
What changes for you
- 01
Pick tasks agents can really do
Score your team’s workflows for value, risk and fit.
- 02
Build a working agent in two days
Connect a model to your tools, data and rules, then test it.
- 03
Deploy with guardrails in place
Human approval, logging and limits before anything goes live.
Included
What you get
A use-case shortlist
Ranked for value and risk.
A working agent prototype
Built on your own workflow.
A seven-part syllabus
From basics to deployment.
A guardrails checklist
Approval, logging and limits.
Session recordings
For the wider team.
Two follow-up check-ins
Help moving to production.
How it works
Three simple steps
- 1
Apply
Tell me about your team, tools and a task to automate.
- 2
Plan
We pick one workflow and prepare safe test data.
- 3
Build and check in
Two days of building, then two check-ins.
Built by your coach
Caspermate
Agents and AI pipelines in production.
I design, build and run AI systems in my own products, including Caspermate. The course is built from that real deployment work.
See CaspermatePricing
One clear price
from$7,500AUD incl. GST
Two days for up to 12 people. Longer build programmes on request.
- Use-case planning call
- Two live build days
- Working agent prototype
- Guardrails checklist
- Session recordings
- Two follow-up check-ins
Syllabus
What you learn and build
Seven parts over two days
- What an AI agent is, and when not to use one
- Choosing and scoring tasks for agents
- Connecting a model to tools, files and data
- Planning, memory and multi-step work
- Testing: making agents reliable
- Guardrails: approval, access and logging
- Deploying, monitoring and improving
Five agents people build first
- Inbox assistant: sorts email, drafts replies and flags what needs you today
- Meeting prep agent: pulls notes, past emails and news into a one-page brief
- Research agent: searches, reads and sums up sources with links you can check
- Lead follow-up workflow: logs new enquiries, drafts a reply and reminds you to call
- Report builder: collects weekly numbers and writes the first draft of the report
Prefer to learn one-on-one?
If you want to build your own agent rather than train a team, 1:1 build coaching covers a personal AI assistant first, then one agent that runs a real business workflow. Six sessions, monthly check-ins after, and no coding needed.
FAQ