01
The Honest LandscapeWhat the current generation of models is actually good at — and what it isn't
We start with a brutal, demo-led tour of Claude, ChatGPT, and two specialist tools. You leave the first session with a calibrated sense of when to reach for AI and when not to.
02
Context, Not PromptsWhy "prompt engineering" is the wrong frame in 2026
The job has shifted. Modern models reward whoever brings the best context — your data, your constraints, your goal, your examples — not whoever knows the cleverest words. You build a personal context library against real tasks from your week. Output: 10 patterns you'll still use in a year.
03
Mapping Your Busy WorkThe repeating admin, the low-leverage tasks, the things eating your Monday
We map your week honestly and pull out the three or four repetitions worth killing first. By session end you've automated one — and seen how to spot the rest on your own.
04
Building Your First AgentFrom "I chat with AI" to "AI does the work for me"
Agents are how AI actually does work in 2026 — they read your inbox, draft the reply, update the spreadsheet, ping you only when something needs you. We walk through Claude Skills, ChatGPT Agents, and a no-code builder. You leave with one working agent handling a real task from your week, and the literacy to spec out the next five.
05
Judgment, Risk, and GovernancePrivacy · hallucinations · when not to use AI · how to brief your legal team
The one most "AI courses" skip. We cover data handling, hallucination detection, and how to have a grown-up conversation about AI risk with people who don't share your appetite for it.
06
Shipping It to the TeamRolling your new workflow out beyond yourself
Capstone: present the workflow you're taking back to your team — what it automates, what it connects, what it hands back in time or revenue. Cohort and instructors critique. You leave with something you'll actually use on Monday.