Brisa Do - Senior Data Scientist

Brisa Do

4.9

Select me!

Brisa is available for hire

Hire Now

Brisa Do

4.9

Gender

Male

Address

Hanoi, Vietnam

Main expertise

Data Scientists Machine Learning AI Deep Learning Data Analysts Data Data Visualization Analytics TensorFlow Python Azure Big Data Architects

Introduction

Brisa Do is a seasoned Senior Data Scientist with four years of experience, committed to establishing herself as a leader in the data science field. She brings a wealth of expertise in coding, analytics, and business acumen to tackle and solve real-world challenges. Brisa is dedicated to innovating and excelling in the development of data-driven solutions that enhance business efficiency and make a significant social impact. Utilizing her skills in advanced analytics, machine learning, and robust coding practices, she uncovers insights that drive strategic decision-making and foster a data-centric problem-solving culture. Brisa is passionate about collaborating with cross-functional teams, contributing to projects that expand the possibilities of data, and aspires to be recognized as a visionary who uses data science to make a tangible difference in both corporate and community spheres.

Project list

Customer Churn Prediction and Retention Enhancement - Vietnam

2021 - 2023

Implemented predictive analytics to identify unsatisfied customers early, reducing churn rates by 5% monthly. This initiative resulted in $30,000 monthly cost savings using big data technologies and Tableau.

Enhanced customer retention by reducing churn rates by 5% monthly (over an unspecified period) by deploying predictive analytics for early detection of unsatisfied customers, leading to $30,000 in monthly cost savings through the use of big data technologies and Tableau. Developed and implemented data-driven churn prediction models, achieving 94% precision.

Technologies: Big data technologies, Tableau

Advanced Machine Learning PaaS Development - Vietnam

2021 - 2023

Led the creation of a Platform-as-a-Service (PaaS) using Spark, Zeppelin, and Hive to greatly improve data and model management for analytic projects.

Spearheaded the development of a Platform-as-a-Service (PaaS) utilizing Spark, Zeppelin, and Hive, significantly enhancing data and model management for analytics projects. Onboarded engineers from four major companies, achieving widespread platform adoption and receiving 100% positive feedback from all users. Drove cost savings by reducing license and service expenses, resulting in a total savings of $700,000 annually. Implemented features supporting both tabular and time-series data, catering to a broad range of analytics needs. Enhanced platform capabilities by integrating AutoML, comprehensive monitoring and alerts, and an efficient scheduling system.

Technologies: Spark, Zeppelin, Hive

AI-Powered Accident Detection System - Vietnam

2020 - 2021

Developed an advanced accident detection system using AI and ML models on the Azure ML Platform, achieving a 92% F1-score in recognizing car accidents. The system involved creating secure data ingestion pipelines and performing data analysis using Python and Tableau.

Delivered a sophisticated accident detection system by deploying AI/ML models through the Azure ML Platform. Achieved a 92% F1-score in identifying car accidents over the project duration. Established secure data ingestion pipelines. Analyzed data using Python and Tableau.

Technologies: Azure ML Platform, Python, Tableau

Autonomous Vehicle Project - Vietnam

2020 - 2021

Developed an autonomous vehicle by implementing Deep Reinforcement Learning algorithms, enabling it to successfully navigate and overcome 35 different test scenarios. The project involved simulating LIDAR sensors to improve feature collection during virtual driving sessions.

Trained the autonomous vehicle using Deep Reinforcement Learning techniques. Successfully passed 35 test scenarios during the testing phase. Simulated LIDAR sensors for enhanced feature collection during virtual drives.

Technologies: Deep Reinforcement Learning

Online 3D Bin Packing Optimization - Vietnam

2020 - 2021

Addressed the challenge of the online 3D Bin Packing Problem, achieving an average fill-rate of 79% throughout the project by utilizing Deep Constrained Reinforcement Learning. Enhanced the process by implementing data parallelism for model replication, which accelerated the training phase by 3 weeks. Developed a proof of concept utilizing Blender.

Solved the online 3D Bin Packing Problem with a 79% average fill-rate by: 1. Employing Deep Constrained Reinforcement Learning. 2. Replicating models with data parallelism to accelerate training by 3 weeks. 3. Creating a proof of concept using Blender.

Technologies: Deep Constrained Reinforcement Learning, Blender

Project list

Education

Expertise

Certifications