I am a Data Scientist & AI Engineer who recently graduated with a Master of Science in Data Science from Arizona State University (ASU).

I specialize in bridging the gap between advanced AI, machine learning theory, and production-grade software engineering. I build systems that automate manual compliance reviews, optimize complex supply chain logistics, streaming graphs, and forecast market regimes.

• 🔭 Core Competencies: AI Agents, LLM Orchestration, Graph Data Science, Mathematical Optimization, Scalable ETL Pipelines, and Predictive Modeling.
• 🎓 Education: M.S. in Data Science | Arizona State University (Tempe, AZ)
• 💼 Open to: ML Engineering, Data Science, and AI Operations roles.

Category	Technologies
AI & Machine Learning
Data Orchestration
Infrastructure & Cloud
Frameworks & Languages

AI-Powered Video Ad Compliance Agent

• Automated FTC and YouTube policy review by transcribing video URLs and performing OCR on on-screen text.
• Designed a visual RAG workflow using LangGraph to cross-examine text metadata against vector policy databases, producing structured compliance audit reports.
• Tech Stack: GPT-4o, LangGraph, RAG, Azure Video Indexer, Python, Streamlit.

Interactive Risk Analysis Dashboard & API

• Built an ML pipeline evaluating default probability on 1.37M LendingClub records.
• Created a FastAPI-backend stress-test engine allowing real-time shifts in macro factors (Unemployment, Federal Funds Rate) with calibrated Platt scaling.
• Tech Stack: LightGBM, FastAPI, Streamlit, FRED API, Chart.js, Platt Calibration.

Enterprise Data Warehousing & Time-Series Forecasting Stack

• Modeled an Amazon AIT BizOps analytics pipeline with a DuckDB star schema executing 35 high-performance analytical SQL queries on 1M rows.
• Developed a 7-page dashboard showing inbound freight metrics, anomaly detection, and SARIMAX volume forecasting.
• Tech Stack: DuckDB, Streamlit, Python, SARIMAX, SQL, Data Warehousing.

Real-Time Graph Data Streaming Pipeline

• Created a distributed data pipeline streaming transactional messages to a graph database via Kafka Connect.
• Ran Neo4j Graph Data Science (GDS) workloads (PageRank, BFS) on the streaming graphs for community detection.
• Tech Stack: Apache Kafka, Kafka Connect, Neo4j, Docker, Kubernetes, Minikube.