Claude AI Skills
€ 600
What you'll learn
This course shows you how to use Claude confidently, safely and effectively across your everyday work, from structured prompting in Chat to handing whole pieces of work to Cowork. You move from 'I've tried it a bit' to confident, capable Claude use, with practical patterns you can repeat every week.
What can you do with Claude AI?
- Stop wrestling with prompts and start getting output you can actually use
- Catch errors and inconsistencies across long documents, reports and data before they go out
- Turn scattered files, policies and client packs into searchable Projects Claude works from
- Build reusable Skills that match your voice, your organisation's standards and your workflows
- Delegate whole pieces of work to Cowork and get back finished reports you can send
- Know where to trust the output and where your professional judgement is non-negotiable
Who it's for
Professionals in any sector who want to get real value from Chat and Cowork across reports, proposals, data and day-to-day admin. No technical AI background needed. Just a curious mind.
Course structure
One project, built on a dummy Q3/Q4 sales data set, runs through every session. Session 2 produces the work you'd actually send to people. Session 3 shows the technical engine behind it. Session 4 puts it all to work on your own tasks.
%20(1).png)
Session 1 · 20 January, 12pm to 1pm
Setup, safety and first prompts
A grounding session so you arrive at Session 2 ready to build. We confirm your Claude plan, install the desktop app, check all three tabs (Chat, Cowork, Code) and install the Claude for PowerPoint add-in. We cover tokens and pricing, download your dummy data set and set up your Cowork folder structure.
Session 2 · 27 January, 12pm to 2pm
Get work done - Claude Cowork
You work with a realistic sales data set and produce the deliverables you'd actually send: branded PowerPoint decks, Word documents and email drafts. You set up your workspace, capture your writing voice, build a custom Skill, connect your real email and calendar, and set up an automated morning briefing. You learn every concept by applying it to the data in front of you.
Session 3 · 3 February, 12pm to 1pm
Build the machine - Claude Code
You give Claude project context with a CLAUDE.md file, then run a subagent swarm: three specialists analysing your data from different angles at the same time. You'll see how Code builds an interactive dashboard and connects to live data through MCP. You talk in plain English and Claude writes the code.
Session 4 · 10 February, 12pm to 2pm
Workshop: build something you'll use.
No new teaching. Bring one real task from your current workload, such as a proposal due this week, a report that has stalled or a process you repeat every Monday, and build it out with Claude. You get support from your Course Leader and learn from peers across different sectors and roles. Everyone leaves with at least one high-value output they can use on Monday morning.
"At the start of the first session, I was told I'd build a working dashboard before the two hours were up. I have zero coding skills. Mention the word "code" to me and I glaze over immediately. Not only did I build that dashboard within the session, two days later I sat down on my own and replicated the entire exercise from scratch - no instructor, no hand-holding. That's the measure of how these courses are designed. The content is genuinely practical, the delivery is clear and encouraging, and the results speak for themselves. Worth every penny — and I'll be back for more."
About the instructor
Rowena Hennigan
Rowena is a Course Leader at the AI Institute and a lecturer at TU Dublin, where she teaches future of work skills. At the AI Institute she leads this Claude course, taking professionals from their first prompts to delegating real work to Cowork. She brings close to two decades of helping teams adopt new ways of working, with a clear focus on what holds up in a real working week.
Frequently asked questions
None at all. Our courses are specifically designed for working professionals, executives, and design teams who want practical AI skills, not a computer science lecture.
- If you are taking a Copilot course, basic familiarity with standard Microsoft 365 apps (Word, Excel, Outlook) is all you need.
- For Claude Code, Claude handles the actual coding while you direct it using plain English.
- For industry-specific paths from the Built Environment, we focus entirely on practical workflows, layout options, and project management—never complex programming.
Yes, absolutely. Security and confidentiality are our highest priorities, especially for our AEC and Finance cohorts:
- Data Privacy: Microsoft Copilot and Claude enterprise plans respect your existing corporate IT permissions. The AI only sees data you are already authorized to access, and your inputs are never used to train public models.
- Course Work: Live workshop sessions use anonymized dummy data so you can focus entirely on mastering techniques in a safe environment. You will then be given clear frameworks to confidently apply those exact same prompts to your live, proprietary project files between sessions.
The market is flooded with generic "how to write prompts" videos. Our edge is entirely different:
- Live, Expert Facilitation: You are learning in real-time alongside an executive-level peer group, guided by practitioners from companies actively deploying AI in the real world.
- Always Up-to-Date: The second an AI model drops a new feature, we rebuild our exercises. You are getting a cutting-edge playground, not stale theory.
- Workflow Integration: We don't just teach tools in a vacuum; we show you exactly how they talk to your existing ecosystem - whether that’s integrating Claude with Microsoft 365 or feeding AI-generated concepts downstream into Revit, Rhino, and Photoshop.
We assume zero knowledge and take everyone from 0 to 60. We start at the basics, customising your tool and writing a first prompt that actually returns something useful, then build from there.
- People who arrive more advanced consistently tell us they still pick something up in every session, so we believe we are hitting the right notes.
- You do need a basic level of digital competence: logging in, copying and pasting, moving between windows. We teach the AI, not the computer.
Yes, completely. Governance and ethics are built into every course we deliver rather than bolted on as a closing session.
- What is safe to put into a paid Copilot tenancy and what is not, in the terms your own IT policy uses.
- The safeguards worth having in place before people start using it on live client work.
- Where the output can be trusted, and where sign-off stays with a named human.
Online, live, and led by an instructor. These are not pre-recorded videos. Questions get asked and answered as we go, and our instructors are good at making sure everyone takes part.
- Two screens are worth having: one to watch us on, one to practise on.
- Recordings are provided, usually within 45 minutes of the session ending.
- You keep the session recordings, the slide decks and the workbooks for 60 days after the course finishes.
Several sessions, and deliberately so. We want you to go away and use what you learned between them.
- Run a prompt on a task you were going to do anyway. Try a skill on real work.
- You come back having felt what it is like, which is when the questions get good and the next session lands properly.
- A single long day looks efficient on a calendar and teaches considerably less.
