Free class · Systems and automation
Every download, email, and note asks you the same questions. What is this, where does it go, and what did we decide last time? This class shows you how to hand those decisions to your AI with one inbox, clear homes for your work, and context it reads every time.
Part 1
Email, texts, downloads, links people send you. Each one is another small decision. Where should this go? Let me explain my business again. What did we decide last time? The operating system removes those decisions, so a lot of little choices become one simple habit.
Without a system, finding something means checking downloads, then the cloud, then searching the whole computer. With one capture point, everything new goes into the inbox. The AI decides where it belongs, files it, and writes down what it did. Your first decision never changes.
Your personal life, your real estate business, and each project are separate entities. Each keeps its own context, so you stop regenerating the background every time. The AI already has what it needs for that scope, and nothing more.
The system also frees you from any one tool. When your context, instructions, and decisions live in your own files instead of one chat, you can point whichever AI is best this month at the same folder. It's a front-end build with a back-end benefit. You put in the setup once, and every task after it starts further along.
Part 2
Divide it however makes sense to you. There's no universal folder tree, and you don't have to design one before you start. The AI discovers the homes you already have and maps them. You authorize the locations and answer only what it can't figure out.
Everything new lands in one capture folder: downloads, notes, attachments, ideas. Your only decision at capture time is to drop it there.
A short 10-me/about-me.md: what you do, who you serve, your current priorities, how you sound, and how you like work delivered. No passwords or private client lists.
Each business, client, and project keeps its own facts, sources, and an AGENTS.md with its rules. A client project never inherits another client's private details.
SYSTEM-MAP.md points to where each collection actually lives, on your computer, in Drive, or in iCloud. The AI builds and maintains it. It doesn't move originals out of their established homes.
Every move the AI makes is recorded: the old name and location, the new home, the reason, and whether it actually finished. Unclear items stay in the inbox with one question.
Think of one sentence you type into AI over and over. That belongs in your about-me file. A short example of your own writing does more than any "professional tone" instruction. Keep it short enough to review and update.
Write your about-me file
Help me write 10-me/about-me.md. Ask me five short questions: my role and audience, my current priorities, how I sound, how I like work delivered, and what I want to review myself. Use only my answers. Include one short example in my own voice. Keep it concise, leave unknown facts unknown, and show me the text before saving it.
Tells the AI nothing
seller update FINAL (2).docx
Tells the AI what it is
oak-street-listing__seller-update__showing-feedback__v02.docx
Owner or project, document kind, short description, then a version only when there is one. Add a date at the front only when it's verified. Keep the real extension.
Use lowercase words, hyphens inside a field, and double underscores between fields. "Final" doesn't prove anything was approved, and a filename alone never proves who owns a file. Ask the AI to explain the evidence for each field before it renames anything, and to leave alone any file whose original name has to stay.
Part 3
Capture is deliberately simple: put the item in the inbox. The AI does the rest later. It reads the item, works out which entity and project it belongs to, names it, files it in the existing home, and logs the reason.
Say a document for one client lands with a client who has a similar name. The log shows where it went and why. You correct it, and add the distinction to that client's context so it doesn't happen again.
Anything the AI can't place stays in the inbox with one useful question. You answer it when you have a minute. Nothing gets guessed into the wrong folder.
Ryan runs an afternoon and an evening filing pass. If your tool can schedule, set one up. A saved schedule isn't proof it ran, so check the log.
Once a project has a home, you name the task and the project. The AI uses the map to find that project's context, instructions, and sources, then does the work. You shouldn't have to paste the same background or dictate a folder path every time. Content works the same way: ideas, sources, drafts waiting for review, approved work, and revisions each have a home, and publishing stays a separate step you take. Start with one real project this week.
Give your AI one real project
Start with this real project: [name the project]. Inspect its existing files and AGENTS.md, map its home and sources, ask me only for missing context, and propose the smallest useful setup. Connect it to my inbox and keep the filing record. Then help me complete one actual task using the project's own context and rules. Tell me which files you actually read and which locations you can read or change in this session. Don't claim access or actions you didn't have.
The test is simple. Open a fresh conversation, give the AI a real task in that project, and see whether it can work from the saved context without asking you to rebuild it. If it can, you spend less attention on filing and explaining, and more on the actual work.
Free with a student account
Create a free student portal account to get the workbook, prompts, templates, and downloads for every class.
Create my free account →Already a student? The same link signs you in.
No. It sits on top of the computer and cloud storage you already use. It's a set of folders, short plain-text context files, instructions, a map, and a log. Your AI reads them when you give it the files or authorized access.
Yes, and with other capable tools. The shared part is the information: context files, instructions, the map, and the filing record. Tools don't automatically share chat memory, permissions, or connectors, so ask the AI which files it actually read and what it can change in that session.
No. Keep originals where they already live, whether that's your computer, Google Drive, iCloud, or a brokerage account. The AI maps those homes and points to them. You're changing how new work flows in, not migrating everything in one afternoon.
Clear, ordinary items, with every move logged. Anything unclear stays in the inbox with one question for you. Filing never authorizes sending, publishing, or changing client records. While you learn the system, start with low-complexity, low-liability work.
Your content workflow gets clear homes too: inputs, sources, drafts waiting for review, approved work, and revision requests. The AI drafts from the right sources into the review folder. Approving and publishing stay separate steps that you take.
Systems and automation
Write any recurring job as trigger, middle, exit condition, and iterations, so AI drafts on a schedule and you keep the send button.
Systems and automation
Show AI the screen, document, photo set, or voice note, make it split what it sees from what it can't confirm, then verify.
Systems and automation
Capture, Analyze, Produce, Iterate: four moves that turn one-shot prompting into a system, run on a quiet prospect, a client picture, and your cash.
Library
The whole library of recorded classes, by topic.
Research sources
Project files and instructions can supply reusable context. A project is not general access to every file on your computer.
Projects give Claude a place for relevant knowledge and instructions.
Shortcuts point to originals without copying them. Access still depends on permission to the original.
Folders and aliases support a simple front door into files that stay where they are.
Why selected context and durable notes kept outside a single conversation help an agent work across sessions.
Product access, limits, local listing rules, platform policies, and laws can change. Re-check the linked authority before important or public work.
Keep building
AI Acceleration: Real Estate is a private, self-directed curriculum for practical AI work across content, communication, lead follow-up, CRM, client research, operations, and relationship management. Build it. Review it. Keep what works.
Explore the Skool community ↗Follow Ryan + AI Acceleration
No ranking, revenue, closing, or time-saving promise. The classes teach the work and the review systems around it.