Thanh Tu Nguyen

As a Freelance Senior Machine Learning Engineer at Carrefour, I’ve reduced ML project deployment timelines from 3 weeks to 1 day by building a Vertex AI-based MLOps package for end-to-end automation, driving a 2-3% revenue increase by developing a real-time deep learning solution using computer vision to detect fraud at self-checkout registers. Additionally, I’ve deployed edge computing solutions in stores to enable real-time insights and faster processing.

My technical stack includes working with state-of-the-art machine learning models in recommendation systems, bandit, and statistical models (winning rate probability or click-through rate), and putting them into deployment at scale with modern tech frameworks to serve our very large base of users.

Before I was a senior machine learning engineer at Voodoo and data lab team of BNP Paribas, the largest bank in France.

At Voodoo, a unicorn tech company that creates mobile games and apps with more than 6 billion downloads and over 300 million monthly active users, I built machine learning solutions for an Ad Network project that will generate 20% total revenue in 2 years. Simply put, the ad network act as a technical and commercial intermediary between advertisers, who are looking to place their ad in another app and publishers who has many apps running. We’ve built a bidding system that allows app developers to rapidly monetize their apps by leveraging machine learning and predictive algorithms.

At BNP Paribas, I was part of a deployment project for extracting relevant information from documents, which allows the automation of more than 500,000 documents per year in 3 different countries. I was also leading a team that was working on a speech recognition system that automated the manual process which consists of translating audio calls into text and making business insights from that. This system helps to reduce 50% of operational costs and to increase more than 100% response time for the company.