The connected layer of context, data, and automations that lets AI do recurring work inside the Microsoft tools you already own. Installed in three steps by JamesTech LLC, Columbia, Maryland.
By Mark Lucas, founder of JamesTech LLC, Columbia, MD. Roughly thirty years on the Microsoft stack, three of them at Microsoft. Artemis program contractor at NASA.
An AI operating system (AIOS) is a working layer of business context, live data, and automations wrapped around an AI model so it can do recurring work inside the tools a company already owns. The model is the engine. The operating system is everything that lets the engine know your business and act on it.
The difference shows up the first time you ask a question. Ask a plain chatbot about your biggest client and it answers like a stranger who met you five seconds ago. Ask an AI operating system and it answers like a teammate who has been in the room for years: who the client is, what you sold them, what is overdue, what was promised on last Tuesday's call.
JamesTech LLC installs AI operating systems for 5 to 200 person firms in and around Columbia, Maryland, on the Microsoft 365 and Power Platform tools most of those firms already pay for. This page explains what gets installed, in what order, what it costs, and how to tell whether you need one.
A morning brief lands on your phone at 7 AM: today's calendar, overdue invoices, what happened in yesterday's meetings, and what needs your attention. You did not open an app to get it.
Before a client call you say "prep me for Karen" and get the full history, what is pending, and what to raise. Status reports draft themselves from meeting transcripts and tracked hours; you review and send. Revenue by client and outstanding invoices come straight from your accounting system, updated daily.
Task management stops being a list. The system knows your priorities, flags what is overdue, tracks what you are waiting on from other people, and tells you what to do next. Brain-dump a voice note and it sorts the contents into tasks, projects, and replies.
Each layer works on its own and delivers value before the next one starts. Most firms feel the difference after the first two.
1. AI Opportunity Review. About a week. We walk through how work actually moves in your firm, find the manual and repetitive parts, and rank three to five automations by the hours and dollars each gives back. You get a written roadmap you own whether or not we go further. From $2,500, credited in full toward the build.
2. Build. Two to three weeks for a first install: the context layer, one or two data sources, the daily brief, and the first automations from the roadmap. Built on Microsoft 365, Power Automate, Power Apps, Dataverse, and SharePoint, so it lives inside your tenant and your existing licenses.
3. Train. Your team learns to run it and to build the next automation themselves. That is the part most vendors skip, and it is the reason the case study on this site ends with a team member automating a 40-hour task on her own.
After the install, most firms continue on a retainer (from $3,700 a month) that adds two to four automations a month, a monthly plain-English status report, and ownership of the AI roadmap. Some take the roadmap and run. Both are fine.
Three ways to get here. The right one depends on how much of your own time you can spend and whether anyone on the team wants to own it afterward.
| JamesTech install | Do it yourself | Generic AI consultant | |
|---|---|---|---|
| Knows your full business context | Yes, built in the first two sessions | Eventually, if you keep at it | Depends on the engagement |
| Built on tools you already own | Microsoft 365 and Power Platform | Whatever you pick | Often a new platform to buy |
| Time from start to first automation | Two to three weeks | Months of evenings | Varies; strategy decks first |
| Who can maintain it afterward | Your team, trained on real work | You | The consultant |
| Cost to start | Review from $2,500, credited to the build | Your time plus subscriptions | Typically $5,000 and up before anything runs |
| Data stays in your tenant | Yes | Yes, if you set it up that way | Ask |
It fits solo consultants and small firm owners who are the rainmaker, the doer, and the admin at the same time. It fits practice leaders inside larger organizations who run a P&L with 5 to 20 people and no admin support. And it fits professional services firms (law, accounting, financial advice, coaching) where time is the product and every hour on operations is an hour not billed.
What you need: a business that is actually running, some digital tools already in place (email, calendar, some kind of accounting), and 60 to 90 minutes for each setup session. What you do not need: technical skills, a big team, or new software.
It is a poor fit for a startup idea with no operations yet, and for firms that want a chatbot on the website and nothing more. Those are cheaper problems with cheaper answers.
JamesTech LLC is at 10015 Old Columbia Road in Columbia, Maryland, and works with firms across Howard County and the Baltimore-Washington corridor in person, and with firms anywhere remotely. Founder Mark Lucas has built on the Microsoft stack for roughly thirty years, including three years as a Microsoft employee, and is an Artemis program contractor at NASA.
If you would rather learn to build the first layers yourself, the half-day AI Operating System training class in Columbia does exactly that. If you want someone to own the whole AI program part-time, see the Fractional AI Officer page.
An AI operating system is the connected layer of business context, data, and automations that lets AI do recurring work across the tools a company already uses, instead of one-off chatbot prompts. JamesTech LLC installs one for small businesses in three steps: an AI Opportunity Review that ranks automations by hours and dollars returned, a scoped build on Microsoft 365 and Power Platform, and training so the team runs it themselves.
No. ChatGPT and Copilot are the engines. An AI operating system is what you build around one so it knows your clients, your numbers, and your processes, and can act on them. Without that layer, every conversation starts from zero.
No. The install connects to what you already run: Microsoft 365, Power Platform, your accounting system, your calendar. Nothing moves. The AI reads from those systems and writes back to them through automations you approve.
The Review takes about a week. A first build, with the context layer, one or two data sources, and the first two to four automations, typically runs two to three weeks. Automations keep landing after that at two to four per month on a retainer.
The AI Opportunity Review starts at $2,500 and is credited in full toward the build. Builds are scoped from the roadmap, so you see the price before any work starts. Ongoing retainers start at $3,700 a month.
Your data stays in your Microsoft tenant and your own systems. Mark has handled Controlled Unclassified Information for NASA and delivered for the DOJ and NATO. Access, retention, and compliance constraints are documented in the roadmap and flagged on every recommendation.
The AI Opportunity Review is the first step of every install. You come away with a ranked roadmap you keep either way.
Book your AI Opportunity Review