I’m a product leader and developer working across enterprise software, healthcare, innovation, and AI. My work spans product strategy, user experience, system design, and evaluation: taking an idea from an early prototype into a workflow people can use.
I focus on the decisions that make AI dependable. Which work is worth automating? What can the system access or change? How will we know when it fails, and how will someone recover?
I also build small tools and write about what I learn. LLM-Expect is my Python SDK for repeatable evaluation. My MCP evaluation case study describes how I tested an agent’s tool choices, arguments, and response to denied access.
Read my product principles or all essays. You can also find me on LinkedIn, GitHub, and X.
For speaking, advisory, product leadership, or AI product work, send me a note.