AI systems for operators
Start here if you want the practical operating layer: how AI agents keep work moving, where human judgment still matters, and what a small business AI system should look like before it becomes another dashboard to babysit.
Field notes on AI systems, operator leverage, acquisition workflows, real estate judgment, and the stack that keeps work moving without constant manual prompting.
Start here if you want the practical operating layer: how AI agents keep work moving, where human judgment still matters, and what a small business AI system should look like before it becomes another dashboard to babysit.
What I automate with AI in the first 90 days after closing on a small business, what stays human, and the exact checklist I run so integration does not quietly erode the deal I just underwrote.
Underwriting is where most investors either burn their evenings or start trusting a spreadsheet they never checked. AI changed how fast I screen deals, but it changed almost nothing about who decides. Here is the working split, workflow by workflow.
Most RV parks never hit a listing site. They sell over a kitchen table to whoever the owner already knows and trusts. That means the buyer who wins is rarely the one with the most capital, it is the one with the most owner conversations in motion. This is the exact five-layer system I would use, and largely do use, to run RV park deal flow with AI doing the memory, lists, and drafts while a human does every real conversation.
Most off-market deal advice tells you to send more letters and make more calls. That advice is right and incomplete. The bottleneck was never knowing what to do, it was doing it every week without dropping anything. Here is the AI deal flow system I actually run: how the lists get built, how owners get contacted, how AI drafts every follow-up, what the screening rules kill automatically, and the three decisions I never let software make.
The real question about AI and your career is not whether it takes your job. It is which side of AI you are standing on. As an employee, AI is the capability that can remove your income in one decision. As an owner, it is leverage you deploy yourself.
Everyone asks me the same question once they hear I buy RV parks: are they actually a good investment, or is this a trend dressed up as an asset class. Here is the honest answer, built from underwriting real parks in 2026, not from a webinar selling you a course.
Most investors lose off-market deals in the silence between conversations, not in the negotiation. The owner says 'maybe next year,' the investor forgets to circle back, and eighteen months later the property sells to whoever happened to call that week. This is the follow-up system I run with AI handling the memory and the drafts while I handle every real conversation, plus the cadence rules and the checklist to build your own.
Most people hear cost segregation and think it is an accounting trick. On an RV park it is closer to a structural advantage, because so much of what you buy is not the building. Here is the practical version, built from underwriting parks as an active buyer, plus where I let AI do the sorting and where my CPA stays in charge.
Every AI implementation quote leads with a subscription price and stops there. The subscription is rarely the real cost. Here is the honest four-bucket breakdown I use across my own businesses, realistic monthly ranges for a solo operator versus a small team, and the budget-capped pilot that tells you the truth in 90 days instead of guessing for a year.
Everyone on the internet is an AI implementation expert right now. Almost none of them have run an AI system through a bad week. Here is the practical definition of real expertise, how to verify it before you pay for it, and the path to building it inside your own business.
AI real estate valuation tools promise instant answers on what a property or a business is worth. The honest version is narrower: they are fast at assembly, weak at context, and dangerous the moment an owner treats an output as a decision instead of a starting point. Here is how I actually use them.
Most people hire an AI implementation advisor expecting a technology decision. What they actually need is someone who can draw one line correctly: what a machine should own and what stays with the owner. Here is the practical version of that job, built from running it across three businesses and a live acquisition pipeline.
Most small businesses do not fail at AI because of the technology. They fail because they have no sequence. Here is the 12 month roadmap I run across my own companies: four phases, one workflow at a time, with a quarterly review that keeps the whole thing honest.
A practical, operator-tested definition of AI integration in business: where it earns its keep, where judgment has to stay human, and the exact checklist I run across a lending business, an AI product, and multifamily acquisitions.
The exact AI-assisted screen I run on local service businesses before I spend a single hour on seller calls or diligence: what AI gathers, the scorecard, the red flags, and where judgment stays human.
Most founders start content to get free traffic and accidentally hire themselves into a new role. The fix is not more discipline. It is an AI content operating system with standards, proof, and a clear owner decision layer.
The useful workflow is not tool worship. It is turning Semrush signals into a repeatable weekly loop: what to write, what to refresh, what to link, and what to measure.
Most businesses start AI integration in the wrong place. Here is the operator's sequence that actually sticks: start with the workflow that burns the most attention, build the smallest useful version, and expand only after it proves itself.
Small scattered deals eat operator attention. Here is my 24+ unit buy box, the AI screening pass every deal goes through before it reaches me, and the go/no-go calls I refuse to automate.
I almost pushed a raw API key to a public repo. Scraped keys run up four-figure bills before sunrise. Here is the 15 minute fix you only do once: no key ever lives in your code again.
Search is splitting between blue links and AI answers that cite sources. How founders earn citations: direct answers, real credentials, current dates, and proof.
AI is the first business expense where the bill shows up after the work is done and can swing 10x without warning. Here are the real numbers by stage, what drives cost, and the 15 minute review that keeps it honest.
The 15-minute daily SEO loop I run with AI: what the system checks, what it drafts, what the founder approves, and how to log proof.
A public build-in-progress: the AI SEO system I am wiring up for my own website, what it automates daily, and what I refuse to hand over.
A practical, owner-level guide to choosing and running AI without turning the business into a tool-management project.
A practical operator guide to AI integration for small businesses that need workflow improvement, clear ownership, and proof before adding more tools.
A practical AI workflow ownership map for founder-led businesses that need automation without losing judgment, customer trust, or accountability.
AI Implementation Systems for Founder-Led Businesses explains a practical operator framework for applying AI with clear standards, useful proof, and business-specific workflow design.
Integrating AI into Business: A Practical Operator's Guide explains a practical operator framework for applying AI with clear standards, useful proof, and business-specific workflow design.
AI Business Integration: A Practical Operator's Guide explains a practical operator framework for applying AI with clear standards, useful proof, and business-specific workflow design.
A prompt can make the first draft faster. A proof loop makes the system trustworthy enough to use every day.
A practical, owner-level guide to choosing and running AI without turning the business into a tool-management project.
If I could only keep one thing from my entire AI setup, it would be a markdown file. Here is exactly what goes in it and how it changes every session you run.
Human-in-the-loop AI is the difference between useful automation and uncontrolled delegation.
A simple operating roadmap for owners who want AI in the business without turning the business into an experiment.
Most AI automation failures are workflow failures, not model failures. Here is what to avoid before you build.
AI can be a strong decision-support layer in real estate, but it becomes dangerous the moment operators let it act like the investor. Here is where I draw the line.
The best AI use case in multifamily is not magic underwriting. It is turning scattered local signals into a cleaner acquisition queue so you can spend more time on the right owners and fewer hours sorting noise.
Zoe is not a chatbot. She is the AI operating layer behind my real estate deal flow, owner-call prep, inboxes, outreach, content systems, CRM checks, Notion, Obsidian, Telegram, iMessage, Codex, Claude Code, and my Mac mini.
My AI team is not a novelty. It is the operating layer behind real estate deal flow, underwriting support, content, inboxes, outreach, and portfolio operations.
Paying cash can feel safe, but seller financing may be stronger when it preserves capital, protects the structure, and lets the seller win too.
Everyone tells you to automate everything you can. That advice is incomplete. The more useful question is its inverse: what should you never hand to AI? Once you draw that line clearly, everything else becomes an automation candidate and the guilt about stepping away disappears.
The first-pass underwriting checklist I would use before spending serious time on an RV park acquisition.
A practical RV park seller financing framework: why seller notes show up, which terms matter, where risk hides, and when creative structure beats paying cash.
A practical RV park buy box for a first acquisition, including price range, site count, occupancy, seller financing, NOI, and red flags.
A plain-English look at why bonus depreciation makes RV parks interesting in 2026, and the tax questions I would confirm before closing.
Most business owners trying to integrate AI end up with a pile of tools, a few partial automations, and a growing list of things to fix later. An AI implementation consultant is not a vendor. It is the strategic operator who turns your AI investments into systems that actually run.
Most small business AI projects fail because owners start with tools instead of workflows. Here is the operator framework I use across three businesses.
The question I get from almost every business owner I talk to right now is some version of: should I hire for this role or try to automate it? I have built AI-native operations across three brands with a combined team smaller than most five-person startups. Here is exactly how I think through that decision.
I run Claude Code across 16 active conversations on any given workday. One iTerm crash used to mean 30 minutes of finding which session UUID belonged to which project, before I could get back to actual work. Now it is one command. Here is how it works and how to install it.
I spent four hours one Tuesday afternoon debugging a single AI workflow instead of calling a lead I had been meaning to follow up with for two weeks. That afternoon taught me more about AI's real time cost than any case study I have ever read.
After building multi-agent AI pipelines across three brands, I can tell you the honest truth: AI does not run itself. The business owners winning right now are the ones who treat human expertise as the steering wheel, not the backup plan.
What I look for in RV park investing in 2026: bonus depreciation, seller financing, cash flow, REP status, and the buy-box filters I use before pursuing a deal.
Before closing on a downtown Charleston STR, I underwrote five specific things that most buyers miss. Here is how I evaluated a short-term rental in Charleston's STR overlay district, why the review history was the real asset, and what surprised me after closing.
After scaling a marketing agency to 7 figures and a 15-person team, two client calls about AI-powered Meta ads tools forced me to confront how quickly the old model was already changing.
Q1 2026 made one thing clear: AI is no longer a side experiment. Companies that do not build AI systems and agents into sales, marketing, and operations will fall behind.