AI Model Evaluation and Training – Image Generation Model

Overview

I worked as a specialist trainer on a contracted image generation model project for a major AI product (via Handshake AI). I evaluated model outputs against criteria set by the client’s research and product teams, flagging edge cases and gaps between intended use and real world behavior across a wide range of use cases. This work required adapting evaluation criteria across shifting product intents and communicating findings clearly to cross functional teams.

What this involved:

→ Extreme attention to visual detail

→ Edge-case/pattern identification across real world use cases

→ Rubric based evaluation

→ Cross-functional feedback communication

→ Evaluation at consumer product scale

I am happy to discuss more about my experience and takeaways in general terms. Reach out!

---