In partnership with
Ok, I know… really? Here comes the hype!
Well, last week I wrote about how Clawbot is going to change things. I think we are literally just scratching the surface here, but I had put this concept into practical use.
About a month ago, I bought a Mac mini and created my first full-time agent. I called her Ally. I landed on that name because it’s a play on my tech company’s name Alliance. (I thought it was clever. 😏)
Anyway, I set a goal. Build an AI team that I can use to impart all of my 20 years of business building and IT knowledge on, and have them monetize it. Basically, I am trying to create a little business machine that can execute all of my crazy ideas and bring value to the world. These are things that I wouldn’t normally have time to do OR have the skills to execute. But my AI team… well, that’s where they shine!
Fast forward through the build process and what I have now is a core team of 8 agents that report directly to my AI Chief of Staff, Ally. She is responsible for managing all the agents, and she reports directly to me. I use telegram to chat with her and I have specific department channels where I can dive deep into a specific subject.
Telegram Chat w/Ally
Building Ally and this team is a proof-of-concept for me. I can’t advise others on AI if I don’t practically learn it myself. I am testing different models, I am stretching the autonomy of the system, I am optimizing, and I am logging lessons learned.
Right now, this tiny Mac that sits on my desk is air-gapped to anything of importance. I don’t trust it with sensitive information yet. I haven’t tested the bounds of guardrails or how it handles vague requests. She’s an experiment at this point.
What I am learning.
Ally has really only been effective for about 2-weeks. There was a decent learning curve in the beginning, but the more information you train her on, the better she becomes.
Ally has her own accounts to critical applications that make her feel like a real employee. She can update Notion, build me a presentation, publish a google doc, write an email, build a website, and much more. I have tuned her memory so that she really is starting to understand my preferences, so much so, that she suggests when I should reconsider or pivot from an idea. I have built in custom indexes that allow me to monitor her intuition, and trustworthiness. I even created a Christian index where we both evaluate her decision making against the moral principles of the Bible. I figure few would do this, but I don’t want to accidentally create the Terminator.
More time with Ally should tell me whether this tech is really going to consume jobs, or if it is merely going to eliminate redundancy and waste. My initial assessment is that it might be a little of both.
Beyond the hype, I think this setup can be an incredible tool for someone. What I do know, however, the setup I have isn’t ready for autonomy or enterprise integration yet. There are too many security implications. However, if you have experience or unique knowledge that can’t be scraped from the internet, I would get in now and start tinkering because this might be perfect for you!
Until next week,
—Jared
Text Me: 314.806.3912
How To Support Me
Let me send this directly to your inbox every week, subscribers also receive special events and downloads that I don’t publish anywhere else. I enjoy your feedback and respond to every email.
A comprehensive guide for addressing the tax talent crisis
A labor shortage in tax is driving the need for a new skill set: one that blends technical tax knowledge with digital fluency.
Automation, AI and data-driven insights now define the role of tax professionals.
This new era of tax is not simply about adopting new tools, it’s about reshaping the skill set and mindset required to thrive in this field. Check out this guide for actionable insights into how to cultivate these skills with your team. See how advanced technologies can help bridge the tax tech gap to increase efficiency, ensure compliance, and drive better decision-making.
