I work from strategy to build: sitting with the business to match the right problems to the right emerging tech, then with the technical minds who prototype it, prove the value, and scale what earns it.
For the first decade of my career, I kept finding myself in the same place: between the people who understood the business and the people who could build for it.
I wasn't a data scientist. I didn't try to be. Instead, I learned enough about both worlds to make them productive together. I called it being an analytics translator. In 2018, McKinsey published the term in Harvard Business Review as "the new must-have role." I was already living it.
What made it real wasn't the label — it was a habit I've carried at every stage: deliberately investing in the adjacent competency, the layer that wasn't in my job description but that made the core contribution more powerful. As an analyst, that meant genuine fluency in data science methods and the business domain. Now, it means hands-on prototyping of AI capabilities without an engineering title. The job description has never been the ceiling.
AI has changed the shape of the gap, not the instinct. The translation layer between domain expertise and technical execution is collapsing. The scarce resource isn't execution anymore. It's judgment — knowing which problems are worth solving, which are the right shape for AI, and how to build toward them without losing what the business actually needs.
That's the work I do now. In 2025, I joined the innovation lab at a Fortune 100 P&C insurer, where I prototype agentic AI and GenAI solutions for the property and casualty insurance domain. I'm not an engineer. But I've invested in understanding how these systems work from the inside — because meaningful dialogue with builders requires it. A colleague recently described me as someone who can engage the non-technical big picture and dive into the technical AI weeds with equal fluency, in the same conversation. That range is something I've built on purpose.
What grounds all of it is something harder to name: a practiced ability to imagine being the person on the other end. The end user. The business stakeholder. The person who will actually use this. Good AI work requires knowing what's technically possible and what a real person actually needs — and holding both at once.
That's the gap I've always lived in. It just keeps moving.