Hi, I’m Karine 👋

data science leader building AI/ML products

With 10+ years of coding experience and 5+ years of industry experience at L’Oréal and J&J, I’ve worked across the full lifecycle: from supply chain optimization to consumer-facing intelligence.

About Me

I build data science and ML systems end-to-end, from predictive modelling to production-ready LLM tools. Always learning, always building :)

Recent Projects

LLM Enterprise Chatbot

AI knowledge chatbot built with an agentic MCP & RAG architecture. Designed to efficiently surface accurate, context-aware answers from L'Oreal's enterprise knowledge base

Tools: MCP RAG Claude, Gemini, OpenAI models GCP

AI/ML Demand Forecasting

Built AI/ML demand forecasting platform to optimise inventory production - using key demand drivers including historical data, future pricing & promotion plans, market trends & more

Won L'Oreal's Global Tech Blazer Award

Tools: Python PyTorch Google Cloud Platform MLflow

Recommendation Engine

Developed personalised recommendation engine for next-best-action commercial insights, increasing sales force efficiency & generating sales uplift

Tools: AWS Cloud Kubernetes Docker

Education

Georgia Institute of Technology

Master of Science in Computer Science

Specialization in Machine Learning – GPA: 3.5/4

University of Singapore

BBA, specialization in Business Analytics

First Class Honours - GPA: 4.6/5
Achieved 3 Dean’s List awards (2018, 2019, 2020), awarded to top 5% of cohort

Skills

Here are some tools that I've used most frequently in live, production pipelines:

LangChain
TensorFlow
PyTorch
Scikit-learn
Python
A/B testing
RAG
Fine-tuning
MCP
SQL
GCP - Vertex Ai, BigQuery
AWS - Elasticsearch, S3, RedShift
Airflow
Kubernetes
Docker
MLflow