Abdullah Ibn Mahdi Abtahi

Deriv Defensor is an agentic co-pilot that predicts which affiliate partners are about to churn, explains why they are at risk, and then launches personalized, automated interventions to protect millions in recurring revenue before it disappears.

Time-aware machine learning model to predict combined cycle power plant energy output, with CRISP-DM workflow, model comparison, and executive-style reporting.

AI design co-pilot that connects to Figma, understands a team's design system, and delivers expert-grounded UX critique (citing WCAG, Gestalt, Nielsen heuristics, etc.). Built for the Gemini Live Challenge hackathon.