Applied AI Engineer
I build AI systems that survive in production.
I spent four years working with data across ecommerce, retail, live events, and nonprofits, where I learned to question the data before trusting the answer. Now I bring that same thinking to AI — understanding the problem first, then deciding what should be handled by business rules, models, or an LLM.
Background
I started as a data analyst. Over four years, I worked across ecommerce, retail, live events, and nonprofits, where I learned that getting the analysis right was only part of the problem. You could find the pattern, build the dashboard, make the recommendation — and still have nothing actually change.
That gap between insight and action is what pulled me into AI engineering.
I wanted to build the systems that could do something with the intelligence we were producing.
Today, I build AI systems end to end, from architecture and retrieval to evaluation and deployment. I care about the engineering around the model — making systems understandable, measurable, and practical to run.
How I think about AI systems
If a rule can solve the problem reliably, I don't add a model just because I can. If prediction is enough, I don't add an agent. The architecture should follow the problem.
An agent shouldn't have unlimited freedom just because it can use tools. I think about what it can access, what it can change, when it should stop, and what happens when it gets something wrong.
I care about how the system behaves when retrieval is weak, evidence conflicts, the model hallucinates, or the cost gets too high. Evaluation, observability and failure handling belong in the system — not after it.
Work
Contact
I am open to AI engineering roles and selective consulting engagements. If you are building something serious, I would like to hear about it.
Open to