Two ways I help teams put AI to work.
One is building: taking an idea from strategy to a system that ships. The other is teaching: giving a team the judgment to use AI well after I leave. Both come from the same place, twenty years of shipping real software, now pointed at AI.
For founders and teams adopting AI who want it done with judgment: a partner who owns the outcome, not a demo that falls over in week two.
Service 01
Building with AI
From product strategy to a shipped system. I write the requirements, direct the build, and orchestrate AI to move fast without losing the plot. You get someone who decides what to build and stays accountable for whether it works.
Decide what to build
We start with the problem, not the tech. I turn a fuzzy goal into a clear spec: who it serves, what it does, where it must not fail, and what a good version looks like.
Build it right, fast
I orchestrate AI across product, engineering, and technical direction, reviewing every output the way I would review a junior's work. Speed from the tools, judgment from me.
Put it in the world
Working software, deployed, documented, and handed over so your team can run it. No black box, no dependency on me to keep the lights on.
What you walk away with
- A written spec you actually own
- A shipped, working system
- Clean, documented handover
- Technical decisions explained
- A realistic build timeline
- One person accountable throughout
Service 02
Fluency training
Hands-on programs that teach a team to use AI with judgment rather than tools: how to set a task up, own the output, validate before trusting it, and know where the tools quietly fail. The full thinking behind it is on the fluency page.
See how the team works
I look at how your people already use AI, where it helps, and where it is quietly creating risk. The program is shaped to real workflows, not a generic slide deck.
Foundations, then tracks
Everyone learns the four habits of mind. Then tracks tailored to how engineering, marketing, leadership, and finance teams actually work, using their own tasks as the material.
Make it stick
Practical workflows, a verification discipline, and habits the team keeps using after I leave. The goal is capability that outlasts whichever model is current.
What you walk away with
- A team that briefs models well
- A verification habit for high-stakes work
- Department tracks built from real tasks
- A shared map of where AI fails
- Reference material they keep
- Accountability that scales with stakes
How it runs
Straightforward to work with.
Start with a conversation
A short call to understand the goal and whether I am the right person for it. If I am not, I will say so and point you somewhere better.
Scoped, not open-ended
Building work is scoped to a defined outcome. Training is scoped to a team and a timeframe. You know what you are getting and what it costs before we begin.
One person, accountable
You work directly with me throughout. No handoffs to a junior you never met, no account manager between you and the work.
Track record
Two decades of shipping, now pointed at AI.
Before this practice, I spent twenty years building web-based infrastructure: the systems companies run on, the parts that have to work. That background is why the AI work stays grounded. I have shipped enough real software to know the difference between a demo and a system.
The full track record is on LinkedIn.
Building something, or teaching a team?
Tell me what you are trying to do with AI. If I can help, I will; if someone else is a better fit, I will point you to them.
Get in touch Read the fluency overview