I build trustworthy machine learning and applied AI systems for anomaly detection, forecasting, retrieval-augmented generation, and decision support.
My work combines statistical modeling, research engineering, data systems, model evaluation, and deployment.
- Certified anomaly detection for IoT and cybersecurity
- Retrieval-Augmented Generation and GenAIOps
- Machine learning forecasting and optimization
- Model evaluation, explainability, and reproducibility
- Data engineering and decision-support systems
A retrieval-augmented AI assistant embedded inside my portfolio, allowing recruiters, collaborators, and technical reviewers to explore my work through natural-language questions.
- FastAPI backend deployed on Azure App Service
- Curated Markdown and YAML knowledge base
- OpenAI embeddings and grounded response generation
- 46 knowledge-base documents
- 237 retrieval chunks
- 277 alias test queries
- 17 evaluation suites
- 100% top-1 and top-k retrieval in the latest audit
Explore Project ARIEL · Open the live portfolio
Co-authored Negative-Binomial and Poisson concentration-inequality frameworks for certified, real-time anomaly detection.
- NB-CI achieved 1.000 AUC
- Realized benign false-positive rate of 0.000 in certified mode
- Approximately 448 microseconds per 10 ms traffic window
- Cross-testbed validation on Gotham 2025
- Manuscripts submitted for peer review
Explore NB-CI · Explore Poisson-CI
Led predictive modeling for a first-place energy analytics project covering more than 1,600 commercial buildings.
- XGBoost forecasting pipeline
- 0.9804 R² on future-period evaluation
- SHAP explainability and residual diagnostics
- Estimated $858K to $1.7M in potential annual savings
Developed a capacity-constrained Python matching system for coordinating institutional needs, contributor capabilities, reciprocity, qualifications, trust, and allocation limits.
- 19 participating institutions
- 194 cataloging tasks
- 565 scored candidate matches
- Reproducible and anonymized evaluation workflow
Languages: Python, SQL, R, SAS, JavaScript
Machine Learning: Scikit-learn, XGBoost, TensorFlow, Keras, SHAP
Applied AI: RAG, OpenAI APIs, embeddings, semantic retrieval, prompt engineering
Engineering: FastAPI, Azure App Service, Git, GitHub, ETL, REST APIs
Analytics: Power BI, Tableau, Alteryx, statistical modeling, model evaluation
- Microsoft Certified: Machine Learning Operations Engineer Associate
- Microsoft Certified: Power BI Data Analyst Associate
- Microsoft Certified: Azure Fundamentals
I am open to Data Scientist, Machine Learning Engineer, Applied AI Engineer, and research-oriented opportunities.