Move from prompting AI assistants to directing agents
Learn how to orchestrate and delegate real work safely and effectively to AI agents in real agentic environments like Rebel, Cowork and Codex in 4 weeks.
Agentic AI AcademyTrusted by over 11,000 professionals from
and moreFrom doer to director, in four weeks.
AI is moving from chat to action. The skill that matters now isn't writing better prompts, it's directing the work: handing real tasks to AI, checking what comes back, and turning one-off wins into workflows that run again and again.
This 4-week program is for people who already use AI and want to move from doer to director. You work on your own recurring tasks throughout, and leave with a workflow you've built, supervised, and can keep using.
Made for people who already use AI.
This program is for people who already use AI in their work. You should be comfortable using at least one AI assistant, understand basic prompting and data-safety, and have one recurring workflow you'd like to improve.
New to using AI? Start with AI Competency.
Built to fit around a full-time job.
Designed to fit around a full-time job, fully remote.
One 1-hour live demo each week, plus 1.5 to 2 hrs of applied practice between sessions.
Across the whole program, including capstone work.
Tool-agnostic foundations that carry across the AI tools you already use.
Build inside Rebel, Cowork or Codex, truly practical.
Issued on completion, yours to share.
Five steps to one working workflow.
Over four weeks you build up to one real, working workflow: mapped, specified, supervised, evaluated, and running on your own work.
Setup
Get connected to your environment (Rebel, Cowork or Codex): sign in, credits, first conversation, email access.
From assistant to agent
The shift from completing tasks yourself to directing an agentic teammate. Connectors, skills and folders; map your current work and pick the tasks worth delegating; build your first reusable skill; Security 101: what AI should and shouldn't access.
Workflows & managing teams of agents
Move from vague prompts to clear task specs (goals, inputs, constraints, output formats, review rules). Run multi-step work asynchronously, supervise what comes back, and judge quality with rubrics. Automations, habits, privacy and multimodal work.
Ecosystem, memory & shared context
How work changes when AI has memory, tools and shared context. Norms and safeguards; the wider ecosystem: no-code builders, browser and computer use, advanced connectors and MCPs, tools like Claude for Excel.
Adopting & championing the new way of working
Demo what you built: the old workflow, the AI-supported version, the controls, and the evidence it works better. Then what comes next: embedding the habit and becoming a go-to person for agentic ways of working in your organization.