Automation · Nashville, Tennessee
The honest one.
This is the newest thing I do and the only one on this site with no client work behind it. I would rather tell you that on the page than let you find out in a meeting.
Where this actually stands. I have not yet shipped an automation for a paying client. I am building the skill deliberately, on my own operations first, because that is the only place I can afford to get it wrong. Everything below is what I can do, not what I have been paid to do.
What automation means, minus the noise
Most of what gets sold as AI automation is a job somebody currently does by hand, on a schedule, from a spreadsheet or an inbox, and does not enjoy. Copying enquiries into a tracker. Chasing a document that never came back. Writing the same six replies. Building the same report every Monday. The work is real and the boredom is measurable.
It is not magic and it is not a robot that runs your business. It is a script that does one specific thing reliably, plus a decision about what happens when it fails, which is the part people skip and the part that matters.
What I have actually built
For my own operations, not for clients: document generation that produces booking contracts from a template, inbox triage that sorts and labels incoming mail by what it needs, and scheduled jobs running on cloud infrastructure that gather data and push it into a sheet without anybody opening a browser.
That is a genuine foundation and it is also not the same thing as having done it for a business with its own systems, its own edge cases and its own tolerance for failure. Those are the hard parts, and I have not met them at client scale yet.
What I would take on
- A repetitive task you can describe in one sentence and currently do by hand
- Something with a clear input and a clear output, where I can tell whether it worked
- Work where a mistake is recoverable, not where a mistake costs you a customer
What I would turn down, at least for now: anything touching payments, anything where a silent failure would go unnoticed for a week, and anything sold on the promise that it replaces a person.
What it costs
Open, and deliberately so. Because this is a capability I am building rather than a product I have priced a dozen times, the rate is discussible and the scope gets agreed in writing before anything starts. If the honest answer to your problem is that it is not worth automating yet, I will tell you that instead of quoting it.
If you want to be the first, there is an obvious trade available: a lower rate in exchange for letting me write it up as a case study. Say so when you get in touch.