AI Foundations 01 · recorded course lesson
How do I delegate a useful job to AI?
Choose a small job you understand, give AI the relevant facts and a clear output, then inspect the result. Finish with a reusable brief and a reviewed draft, rather than a longer conversation with no usable result.
Ryan Wanner · published January 2026
All classes and lessonsWatch the change in mindset
6:47 course lesson · Watch with captions on YouTube
Every class on this page is free to read and watch. No account needed.
Turn “help me follow up” into a job
A useful brief names the reader, facts, output and checks. Work with the fictional case first, then replace it with appropriate facts from a job you understand.
Edits stay in this tab. Copy the version you want to keep.
Work through the brief
Name one job
Choose a recurring task you already understand. Name the reader and the useful next action. Keep the first job small enough to review.
Separate facts from questions
Use the supplied fictional case or your own permitted notes. Put known facts in one list and missing information in another. Do not fill the gaps with guesses.
Define the deliverable
Ask for a short draft with one scheduling question. State the format, factual limits and review criteria before running it.
Review and revise
Check every factual sentence against the notes. Mark keep, change or remove. Ask for one targeted revision and inspect the result.
Save a reusable brief
Replace case details with named input fields. Keep the acceptance checks and a representative example, with no previous client's details.
What to listen for
searcher to a delegator.
having a human in the loop to validate the outputs.
context over cleverness.
Repair the brief when this happens
Asking for help without defining the job. Name a reader, output and next action so you can judge the result.
Letting the model fill missing facts. Keep the budget, property match and other unknowns explicitly unresolved.
Saving the entire client example. Retain the reusable structure and checks; remove case-specific private details.
Keep the brief and review checklist
Create a free student portal account to download this class's files.
Download the delegation brief and checklistRead the full practice sheet here
# Delegation brief and review checklist Current companion to AI Foundations 01, prepared November 8, 2026. This is a new practice resource, not a recovered original handout. Job: Reader: Useful next action: Source facts and dates: Unknowns: Output format: Limits: Acceptance checks: ## Fictional case Cedar House Realty has permission to follow up about a condo showing. The person requested two bedrooms and a step-free entrance, is available Tuesday after 3 p.m., and has not stated a budget. ## Review the draft - Every factual statement comes from those notes. - No property match, budget or accessibility certification was invented. - One useful scheduling question advances the conversation. - The draft remains unsent. ## Keep the useful part Version/date: Change made: Why it helped: Fields to replace for the next case: Private details removed:
Before your first delegated job
Editorial learner questions.
What should I delegate first?
A small recurring output you already understand and can check against available facts.
Do I need automation software?
No. Start in an ordinary chat with appropriate source notes. Automation is a later decision.
How do I know the result is good?
Define the factual limits and useful next step first. Compare the actual draft with those criteria.
Why is there a correction beside the recording?
The historical opening conflates significantly positive impact with all positive impact. The corrected survey categories do not establish the recording's claimed results gap.
Selected passages from the recording
Selected caption excerpts from the historical recording. These are deliberately non-contiguous passages, not the full transcript. Current practice and corrections are written separately above; the original recording is unchanged.
0:51 · From questions to delegated work
but one of the biggest mindset shifts that you can have is going from a searcher to a delegator. Most agents and other real estate pros treat AI like it's Google. They go, they ask their question, they copy and paste the answer, and they repeat that cycle until they get what they want.
The people who think differently give it the information it needs, give it clear outputs, and then they review the output and they iterate that forward.
3:01 · Fit the work to the business
Now that being said, I know that there are a lot of different ways to operate a real estate business. So the goal is not to just build a system that works for my business and hand it over to you. Ideally, that would be perfect. But we may acquire customers differently. We may operate differently. We may have a different set of principles as it relates to our marketing.
4:18 · Keep human validation
There are a couple core things that you should go back to consistently. Things like having a human in the loop to validate the outputs. Things like being contextoriented instead of just being clever and trying to input a new prompt every time.
5:49 · Context over cleverness
So again, I said it a little bit earlier, but it's context over cleverness.
Build the next skill
Six Mental Models for Judging AI Work
Use source facts and explicit criteria to judge the result.
Revise and Save Useful AI Prompts
Keep a reversible version of a request that works.
Build a Context Card That Stays Useful
Keep relevant facts current and resolve conflicting notes.
After your first brief
Choose the next job from your actual workload.
You have a way to brief a bounded task. If choosing the right task is the obstacle, review how the Agent Constraint Accelerator approaches your business.