Talent and people teams: recruiters, sourcers, talent acquisition and HR business partners, people ops, and the leaders and managers who hire with them. Every exercise is built around real talent and people work: intakes, sourcing, scorecards, debriefs, offers, exit interviews, hiring updates.
Yes. No paywall and no upsell inside the lessons, and it’ll stay free.
No. Everything is done in plain English, including building Skills, Artifacts and an agent in Part 2. Claude writes any code. There’s an optional bonus lesson if you’re curious what’s under the hood.
Part 1 works in any AI tool your company has approved. Part 2 is built on Claude, with notes on the ChatGPT equivalent each day. A paid plan helps, because free tiers run out of messages partway through.
Part 1 (weeks one and two) takes about five minutes a day. Part 2 (weeks three and four) takes 10–20 minutes a day, because you’re building things. Miss a day? Just pick up where you left off.
The last day of the course, when you show your team what your AI can do: the problem, a live demo, the evidence and what to try next, in five minutes. Booking it now gives you a deadline, and a deadline keeps you going. Book your Demo Day.
No. Part 1 covers what most people need to use AI well every day, and earns the Onboarded badge. Part 2 is for people who want to build (Skills, Artifacts and an agent) and earns the Promoted badge.
Please do. , pick the same Demo Day and compare what you build. It’s much easier to keep going together. Want to go further? We also run private workshops for teams.
A .md (Markdown) file is a plain-text file with a few symbols for headings and lists. Every AI tool reads it well, you can open it in any text editor, and it keeps your context portable if you ever switch tools.
A file in your project folder that Claude reads automatically at the start of every conversation there. Think of it as your AI’s handbook: who you are, how your team works and the rules it must follow. You write it on Day 17.
A packaged instruction for a task you do often, like kicking off a new role. It’s a folder with a SKILL.md file inside: a name, a one-line description that tells Claude when to use it, and the steps in plain English. You build your first one on Day 20.
A small app or document Claude builds next to the chat, which you can publish as a link. Under the hood it’s code, but you only describe what you want and test it. You build two in Part 2.
An agent is AI with a job: instructions, tools, a trigger such as a schedule, and the freedom to work through several steps on its own. Yours drafts the work; a person always approves it. You build one on Day 26.
Not in these exercises, and never in a personal AI account. Every lesson comes with a made-up scenario. For real work, use your company’s approved tools and follow its AI and data policies.
It depends on where you hire and how you use it. Some places regulate AI in hiring: the EU AI Act treats recruitment AI as high-risk, and New York City requires bias audits for some automated hiring tools. This course keeps AI in the assistant’s seat: it drafts and organizes, and people decide. Day 27 helps you write your team’s boundaries. Check with your legal team.
In this browser only. We don’t have accounts. If you sign up for the calendar invite, we store your email address, your time zone and your Demo Day date and place, only to send that one email.
The team at Screened, makers of Ryan, an AI recruiter that screens resumes, interviews candidates and keeps them updated, so hiring teams only meet the right candidates. Every talent and people team has been told to “use AI,” but few have been shown how, so we wrote the plan we wished existed.
A small hands-on session where you bring one workflow you repeat every week and build an agent for it, with us in the room. Join an open workshop near you, or book a private one for your team. Register interest and we’ll plan around it.
