Agentic Systems
Learn to build one focused AI-agent workflow.
Build Reliable AI Agents
Work with Sunil Mathew on a bounded use case. Connect the agent’s task, tools and context into a workflow you can build, examine and explain.
| Programme overview | Details |
|---|---|
| Who this is for | Developers, product and programme managers exploring one focused workflow; hands-on scope depends on technical starting point. |
| Learning outcome | Build a bounded AI-agent workflow and explain its tools, behaviour, human review points and limitations. |
| Duration | 3 weeks |
| Format | Live cohort learning, with 4–5 hours of independent work a week in addition to live sessions. |
| Cohort | A live cohort of 8, never more than 10. It starts once 8 participants have confirmed. Sessions are not recorded. |
| Certificate | A certificate of completion from Sunil Mathew. It is not accredited by a university or certification body. |
| Hours | 10 live cohort hours |
| Fee | $499 per participant |
| GST | Not charged at the moment. If that changes, we will tell you before you pay. |
| Next step | Request a fit call |
One workflow, built and understood end to end
A focused workflow gives you room to investigate the decisions. What is the agent responsible for? Which actions are allowed? What information does it need? Where should a person intervene?
You build the core workflow, examine its behaviour and use feedback to identify what needs to change.
Four decisions to work through
- Responsibility
define the task and where the agent's responsibility ends.
- Workflow
connect tools and context to that task.
- Evaluation
examine whether the agent does what you intended and where it fails.
- Limits
explain human review points and the work needed before broader use.
The intended result is a bounded agent build with a clearer account of its workflow and limitations.
Does this fit your starting point?
Developers can discuss a practical build. Product and programme managers can explore the design and evaluation work relevant to their responsibilities. The hands-on scope is matched to your technical starting point before joining.
If your problem needs several agents working together, compare Design and Build Multi-Agent Systems.
Your commitment
10 live cohort hours with Sunil, plus 4–5 hours of independent work per week. The fee is $499 per participant. Scope, schedule and payment arrangements are discussed before joining.
Use the independent time to build, inspect behaviour and prepare questions for the next discussion.
Guidance from Sunil Mathew
Sunil's 26-year engineering career includes past roles at Google, Amazon and Walmart Labs. He brings that experience into practical teaching that asks you to connect a working result to the reasoning behind it.
Can I bring an idea from work?
Yes. Describe it at a high level, and we shape it into something safe to build in a focused learning project. Taking it into your organisation needs further engineering and review.