I lead AI implementation from strategy through adoption, and I build working AI tools myself, so every recommendation is grounded in what the technology can do today.
People feel faster, and the results still lag. The gap shows up in a few familiar places.
Teams got tools and a kickoff session, and usage faded within weeks because nothing about the daily workflow changed.
People produce more, and the extra output piles up in the same review, approval, and handoff steps as before.
Nearly 40% of the time AI saves goes to rework. Clear guidance on where AI fits in each workflow keeps more of that time.
Each team experiments on its own, so wins stay local and risks go unmanaged.
Rework figure: Workday, January 2026
Tools are the easy part to buy. The work is in fitting them to how your teams deliver.
I look at your delivery lifecycle end to end and identify where AI removes real friction and where it would add risk or rework.
A clear strategy with guardrails and success measures earns support from leadership and from the teams doing the work.
We prototype on actual projects with the actual team, and redesign the surrounding process at the same time.
Training, documentation, and adoption tracking turn a successful pilot into the normal way of working.
Employer details stay private. Walkthroughs and demos are available in conversation.
Authored a strategy for bringing AI into design and its delivery partners at a publicly traded HR technology company under a company-wide no-AI policy, and won cross-functional buy-in from legal, risk, product, and development.
Read the case study →Designed and built a tool on the Claude API that scopes incoming design requests, predicts delivery timelines against a team's real skills, and recommends designer pairings.
Demo available on requestAn AI-driven training system that teaches firefighters their department's policies and procedures, adapting to what each learner already knows.
In developmentShare where AI is falling short today, and we will start there.