I build the systems, processes, and infrastructure that let digital-first organizations scale with clarity and precision.

For 10 years I've run the technical and operational engine of a global media company. Four million participants across 180 countries, over a thousand live online events a year, and the infrastructure underneath all of it.
I own the whole stack, which is the unusual part. I configure the platforms and negotiate the contracts that pay for them. I write the automation and manage the people who run it. Most operations leaders do one side of that, and doing both is why I can tell you within a day whether something is a tooling problem, a process problem, or a staffing problem.
In practice that has meant near 100% on-time, on-budget delivery across those launches, hard-dollar savings through license audits and vendor renegotiation, and holding output steady while the team got leaner. When platforms fail at scale, I'd rather find it than hear about it from a customer.
I hold an MBA in Sustainable Enterprise and I've been building with AI since it became useful, shipping customer-facing features and rebuilding production workflows rather than adding subscriptions.
Most organizations bought the tools first and went looking for the problem second. I work the other direction. Every place I've put AI to work started with something that was already costing us time, money, or quality, and the tool was the last decision, not the first.
The result is not a pile of subscriptions. It is a smaller, faster operation that ships things it could not ship before.
I use AI to design, code, and ship features that customers actually touch. Internal ideas that once died in a backlog waiting for engineering capacity now get built and tested. The distance between spotting a gap in the customer experience and doing something about it has collapsed.
Usage reports, spend data, and platform metrics usually sit unread because pulling meaning out of them takes hours nobody has. I use AI to interrogate that data and bring leadership the answer they need. In practice that has meant right-sized licenses, renegotiated contracts, and bandwidth costs brought back under control.
Automation exposes every assumption nobody wrote down. When a workflow resists being automated, it is telling you the process underneath is broken, and fixing that is usually worth more than the automation itself. Document it, clean it up, then hand off the repetitive parts.
Transcription, video and audio editing, chaptering, and metadata across a large and growing content library. Work that used to consume hours of contractor time per asset now moves in a fraction of that, which changes what a lean team can realistically take on.
Accessibility standards and transcript accuracy are not optional, and at volume they are expensive to hold to a high bar by hand. AI-assisted review makes consistent quality achievable across an entire library instead of only on the newest releases.
When a long-tenured person leaves, the real loss is everything that lived in their head and never got written down. I use AI to turn scattered notes, threads, and recordings into documentation the next person can actually work from.
Some decisions need a person. Judgment calls, sensitive communication, and anything where being wrong is expensive and hard to detect. Part of integrating AI well is being clear about where it earns its place and where it quietly introduces risk you will not notice until much later. That line is an operations decision, not a technology one.