Key Skills

Freelancer Overview

I helped launch and support an AI assistant used by more than ten clients, which meant designing and running the data workflow behind the model. I wrote simple labeling guidelines, built small gold-standard sets, and set up a review rhythm so we caught edge cases early. I focused on clean examples, clear intents, and consistent outcomes rather than volume for its own sake. Day to day I tracked quality with spot checks, looked for drift, and kept a short changelog so product, sales, and support saw the same source of truth. What sets me apart is that I connect the labeling work to real user value. I sat in customer calls, turned their language into training examples, and measured success with practical metrics like first-contact resolution and time to answer. I am comfortable coordinating with cross-functional teams, keeping data privacy in mind, and documenting decisions so the next person can reproduce the result. The result is training data that is not just accurate, but useful, because it reflects the problems customers actually want solved.

Labeling Experience

Education

Work History