Summary

Data Scientist with 6+ years of experience leveraging data and machine learning to solve business problems at scale.

At Cake By VPBank, designed end-to-end credit scoring systems that enabled aggressive loan disbursement growth while maintaining target bad debt ratios, directly supporting the bank’s lending expansion strategy.

At Zalo, delivered recommendation systems serving millions of users across multiple products, driving personalized content experiences that significantly improved user engagement and retention metrics.

Contact

Experience

Senior Data Scientist - Cake By VPBank

May 2024 - Present

  • Designed and delivered end-to-end credit scoring systems leveraging bank and partner data, achieving around 40 Gini and enabling loan disbursement growth while keeping bad debt within target.
  • Established reliable and reproducible ML infrastructure with DBT, XGBoost, IaC, MlFlow, real-time monitoring, and alerting to ensure stable model operations in production.

Senior Data Scientist - Zalo (VNG)

Jul 2023 - Apr 2024

  • Delivered an automated music genre tagging system using Transformer and CNN on Mel spectrograms, enabling self-service genre labeling for millions of songs on ZingMP3.
  • Designed and launched a personalized playlist system for millions of ZingMP3 users, increasing homepage CTR from 3% to 8%.
  • Owned end-to-end ML pipeline from data collection and feature engineering to production monitoring.

Data Scientist - Zalo (VNG)

Apr 2022 - Jun 2023

  • Built music feature store for ZingMP3 generating audio attributes (danceability, energy, tempo, valence) for recommendation systems using deep learning (CNN + LSTM).
  • Delivered a recommendation engine for ZingMP3 combining collaborative and content-based methods, extending continue-song list length from 13 to 16 and growing adoption from 5M to 8M users.
  • Launched NLP-driven article recommendation systems for ZingNews (TF-IDF, BM25, NER, POS tagging, word segmentation), increasing views in recommendation sections by 200%.

Data Engineer - Zalo (VNG)

Feb 2021 - Mar 2022

  • Built a lightweight but effective data platform for ZingMP3 enabling A/B testing, real-time analytics, and visualization dashboards using Java, Spark, Parquet, and ReactJS.
  • Empowered product teams to make data-driven decisions with self-service analytics tools.

Software Development Collaborator - Zalo (VNG)

Feb 2020 - Jan 2021

  • Trained and mentored in recommendation systems research, exploring collaborative filtering, content-based, and hybrid approaches.
  • Built demo applications to showcase learned algorithms using ReactJS, TypeScript, Java, MySQL, TensorFlow, Spark, pandas, and scikit-learn.
  • Collaborated with product and engineering teams to convert research into deployable features.

Research Intern - High Performance Computing Lab (HCMUT)

Jun 2019 - Jan 2020

  • Supported implementation of scalable on-demand data aggregation pipelines using Apache Kafka and Apache Nifi.
  • Conducted experiment design and performance measurements for data-intensive systems under faculty guidance.

Education

Ho Chi Minh University of Technology

Bachelor’s Degree in Computer Science (2017 - 2021)

  • GPA: 8.01
  • Relevant coursework: Machine Learning, Computer Vision, Data Mining, Data Structures and Algorithms, Operating Systems, Computer Networks, Parallel Computing.
  • Thesis (9.46/10): Surveyed existing action recognition methods and proposed a Multistream Attention-Enhanced Adaptive Graph Convolutional Neural Network (MS-AAGCN) for skeleton-based action recognition in PyTorch, achieving competitive benchmark performance.

Skills

  • Domain Expertise: Credit Scoring, Recommendation Systems
  • Data Science: XGBoost, CatBoost, LightGBM, scikit-learn, TensorFlow, PyTorch
  • Data Engineering: DBT, Airflow, DataHub, BigQuery, Apache Spark, Hadoop, Parquet
  • Data Analytics: A/B Testing, Metrics Design, Dashboard & Visualization, SQL, pandas, matplotlib
  • Business Analysis: Data-Driven Decision Making, Problem Framing, Experiment Design, Stakeholder Communication
  • MLOps: MlFlow, Feast, Kubernetes, Prometheus, Grafana, Terraform, Docker, Git, CI/CD, IaC, GCP
  • Languages: Python, SQL, Java, Bash, JavaScript, C