AI Engineer & Data Scientist
I am a data and analytics professional with 5+ years of experience spanning compliance, risk assessment, and machine learning. I build end-to-end ML pipelines, design interactive dashboards, and explore the intersection of causal inference, NLP, and AI ethics to deliver impactful, production-ready solutions.
I speak English, Haitian Creole, French, and Spanish — because great insights should travel across languages, not just dashboards.
A browser extension that detects deceptive UI patterns (dark patterns) on websites in real time using a two-tier ML pipeline. A lightweight client-side gatekeeper filters candidate elements before sending them to a cloud-hosted backend for classification. The system includes online retraining with Golden Set evaluation and a performance dashboard to monitor model drift.
Built an ML pipeline to predict California wildfire risk using satellite imagery, weather data, and historical fire records. Engineered geospatial features and trained XGBoost models to forecast fire probability at the county level, enabling proactive resource allocation for emergency response teams.
An interactive Flask website visualizing mental health and substance use trends across the United States using NSDUH public health data. Features Altair-powered interactive charts, geospatial analysis, and narrative storytelling that transforms complex health data into accessible insights.
Automated suspicious activity report (SAR) summarization across multi-tab Excel data using Python and pandas with PII-safe ingestion. Built a companion Power BI dashboard for production tracking, reducing manual review time and improving audit trail completeness for compliance teams.
Analyzed global refugee movement patterns using Neo4j graph analytics to uncover migration routes, host country networks, and temporal trends. Visualized complex relational data to surface insights about displacement flows and resettlement patterns across regions and time periods.
Investigated what financial factors drive credit pricing in peer-to-peer lending using Lending Club data. Built an OLS regression model to analyze borrower characteristics and loan attributes. Discovered that Lending Club's internal credit grade alone explained over 90% of interest rate variance, revealing grade as a near-black-box pricing mechanism.
Used computer vision techniques to explore which visual features drive art-style classification. Incorporated edge detection, HOG, LAB color features, local binary patterns, and ResNet50 embeddings. Achieved up to 73% accuracy with minimal performance loss after PCA-based dimensionality reduction across 7 art movements.
Python, SQL, Pandas, NumPy, LangChain, Hugging Face, spaCy, scikit-learn
Machine Learning, NLP, Deep Learning, Causal Inference, Statistical Analysis, Regression, Hypothesis Testing, Graph Analytics (Neo4j)
Tableau, Power BI, Excel, Altair, Matplotlib, Jupyter, Git, AWS, Docker
SOC 1/2, SOX, IT Audit, Risk Assessment, Controls Testing, EU AI Act, NIST AI RMF
English, Haitian Creole, French, Spanish
Claude Code, Cursor, ChatGPT (GPT-4), Kimi — for code generation, debugging, and accelerating ML workflows
I am passionate about building AI that is not only accurate but also fair, transparent, and human-centered. I love collaborating on data-driven projects and sharing what I learn along the way. If you are working on something at the intersection of data, ethics, and impact, I would love to hear about it.
Say HelloMaster of Information and Data Science (MIDS)
School of Information
biankapaul@ischool.berkeley.edu
Bachelor of Science
Previous undergraduate studies