Fred Mubang

About Fred Mubang

PhD Data and AI Scientist with 8+ years of experience building and deploying production AI/ML systems across large-scale datasets (100M–B+ records), currently at Experian. Specialized in GenAI and modern NLP (LLMs, embeddings, semantic retrieval, RAG), rapid prototyping with rigorous offline evaluation, and productizing models into scalable pipelines using Python, SQL, and distributed compute. Strong track record translating ambiguous business goals into measurable model improvements through experimentation, forecasting, and causal analysis, and partnering with engineering teams to ship reliable ML services.

Experience

1. Machine Learning and Data Scientist

   Experian

   Oct 2022 — Present

   Designed and built an internal LLM-powered chatbot for enterprise employee use, leveraging Python, Transformer-based models, embeddings, and pgvector to enable semantic search and retrieval across internal knowledge sources-Developed GenAI prototypes using large language models to automate data-quality validation, issue triage, and documentation workflows, significantly reducing manual analyst effort and review time-Built and deployed production-grade machine learning models (XGBoost) for mortgage attribute prediction, supporting downstream analytics and decision-making pipelines-Led end-to-end data preparation and feature engineering workflows, including large-scale data collection, cleaning, validation, and transformation for model development and inference.• Improved a large-scale, data-driven contactability algorithm using Hadoop, PySpark, and Python to identify optimal phone and email contact information across ~200 million individuals, achieving a 30% lift in phone prediction accuracy and a 20% lift in email prediction accuracy-Designed a predictive household size model for healthcare Federal Poverty Line estimation using linear regression and SMOTE-based data augmentation, achieving a 40% lift-Operated on datasets ranging from hundreds of millions to billions of records, applying distributed data processing and scalable ML pipelines to support high-impact production use cases-Regularly utilized Python, PySpark, Hadoop, Pandas, NumPy, and scikit-learn for large-scale data analysis, feature pipelines, model training, and evaluation-Owned project execution end-to-end by defining objectives, managing timelines, coordinating cross-functional contributors, and driving projects to successful completion-Produced detailed technical documentation and executive-ready presentations to communicate methodology, results, and business impact to senior leadership and stakeholders.

Education

• University of South Florida

  Post Bachelor Studies in Computer Science

  2017 — 2018

• University of South Florida

  PhD Computer Science, Machine Learning/Artificial Intelligence

  2018

• Berklee College of Music

  Bachelor's degree, Music Business

  2010 — 2014

• Hillsborough College

  Post Bachelor Studies in Computer Science

  2015 — 2017

• University of South Florida

  Master of Science - MS, Computer Science/Machine Learning