Skip to content

AI Adoption

Adopt what makes your team better. Skip what just makes the budget bigger.

Overview

What this is

Most of what gets sold as AI right now is software looking for a problem. I help you sort the real opportunities from the noise, clean up what is already in your stack, and make sure the tools you do adopt make your team better at their jobs instead of just making the budget bigger. Your team is already using AI. The work is doing it on purpose instead of by accident.

The most common mistake I see is treating AI adoption like a software purchase. Evaluate vendors, run a pilot, roll it out. The real work is upstream of the tool. What decisions are you trying to improve, who owns the output, and what does good actually look like. Get that right and the tool selection becomes almost trivial. Get it wrong and you end up in pilot purgatory with a subscription and nothing to show for it.

Capabilities

What I do here

What I do here

  • Inventory the AI tools your team is already using, including the ones nobody approved, and figure out what data each one can reach
  • Sort the workflows where AI genuinely helps from the ones where it quietly degrades quality
  • Build the workflows worth building, with human review where the output actually matters
  • Write a usage policy your team can follow without slowing down
  • Set the guardrails so speed does not turn into a data problem you find out about later

How I Work

How I work

01

Find what is already running

Before recommending anything, I map what your team already uses and what it touches. The picture is always bigger than leadership thinks.

02

Separate signal from noise

Not every workflow should have AI in it. I tell you which ones genuinely benefit and which ones are better left alone, and I am willing to say the honest answer is fewer than you hoped.

03

Build with judgment, not just automation

Automation surfaces the work. A human still decides. I build systems where the machine does the repetitive part and your people keep the judgment.

Judgment

Tested and rejected

The tools I have tested and rejected

Dedicated prompt engineering platforms come across my desk regularly, and most do not survive the review. The pitch is that prompt engineering needs its own tool category. The reality is that the foundation platforms, configured with discipline, cover the workflow without another vendor, another subscription, and another set of credentials to manage. When a tool exists mostly to rent you back capabilities you already own, I say so.

Proof

From the work

01

The math most vendors are betting you will not do

A funded brand came to me ready to sign an enterprise AI contract, per-seat pricing with usage fees and annual escalators. I ran their actual use case first: internal documents and repeatable workflows, nothing that needed a frontier model on tap for every employee. An in-house deployment on their own infrastructure handled it. Full transparency, that build was my contract and I got paid to do it. They still came out tens of thousands of dollars ahead of the subscription over its life, and they own the result instead of renting it. If you want to know whether the tool you are about to buy is the right one, the AI Readiness Audit runs that math before you spend a dollar.

AI Readiness Audit

FAQ

Primarily Anthropic's Claude and the major foundation models, configured for your actual work. The assessment and the governance approach apply no matter which platform you land on.

By building the guardrails into the workflow itself. Data handling rules and review checkpoints live inside the tools your team already uses, so the safe path is also the fast path.

Training teaches people how a tool works. Enablement builds the systems and the judgment layer that make the tool produce consistent, secure results across the whole team, not just for the one person who is good at prompts.

Usually yes, and usually not in the way people expect. The risk is rarely the tool. It is what the tool can reach and what nobody is watching. That is exactly what the AI Readiness Audit is built to find.

You do not need to adopt everything. You need to adopt the right things.

Start with a fixed-scope assessment