As a Freelance Data Scientist, I design, build, and deploy end-to-end data science and AI solutions. For a customer churn project, I built a 10-phase pipeline on 7,043 IBM Telco customers; its XGBoost model achieved 0.86 AUC, and I deployed a five-tab Streamlit dashboard.
I also re-engineered a customer propensity workflow into a distributed PySpark pipeline, orchestrated with Apache Airflow and containerised with Docker. In a separate project, I built a locally hosted NLP assistant using Ollama, with tool use, contextual memory, and task automation.
Earlier, at Brainery Code, I taught robotics and STEM and designed data-driven evaluation frameworks. At Stylar Digital Pte. Ltd. and Pan Pacific Hotels Group, I used analytics, KPI dashboards, and CRM systems to inform operations and loyalty programme decisions.

