📌 I work at the intersection of statistical theory, interpretable machine learning, and real-world clinical data.
Focus: Interpretable ML · Nonparametric Statistics · Clinical & Scientific AI
 Peer-Reviewed Paper MDPI Mathematics, 2025 |
 Open Source Contributor qdrant, llama_index, chroma, langflow |
 Multi-Language Ecosystem PyPI · NuGet · RubyGems · Maven |
 Healthcare & Research NLP, risk models, genomics |
"Models should not only predict well — they should explain well."
I approach modeling through three principles:
- Statistical validity before scale
- Interpretability before optimization
- Domain meaning before deployment
My research interests include:
- interpretable and explainable machine learning (post-hoc & intrinsic)
- permutation-based, resampling, and nonparametric inference
- dimensionality reduction with geometric and statistical intuition
- robustness, stability, and noise-aware modeling
- translating statistical theory into clinically actionable insights
Used for statistical modeling, interpretability research, clinical AI systems, and multi-language package development.
An Model Context Protocol (MCP) server for searching, analyzing, and retrieving academic papers.
- Purpose: Integrates arXiv and Semantic Scholar directly into AI coding assistants (like Claude Code/Desktop).
- Features: Page-level text extraction from PDFs using PyMuPDF (fitz), citation graph traversal, and advanced search filters.
- Tech: FastMCP, Python, HTTPX, PyMuPDF.
Nonparametric Combination (NPC) and bootstrap-based risk stratification model.
- Purpose: Reproducible statistical analysis framework for our peer-reviewed research in Necrotizing Fasciitis.
- Published: MDPI Mathematics, 2025
- Tech: Python, NumPy, Pandas, Scipy.
End-to-end clinical NLP platform for medical entity extraction, clinical sentiment analysis, topic modeling, and automated ICD coding.
- Purpose: Privacy-preserving processing and deep learning pipelines for unstructured health records.
- Tech: Python, PyTorch, Transformers, FastAPI.
Scalable clustering framework for big data using KMeans++, DBSCAN, BIRCH, OPTICS and DENCLUE.
- Applied to: NYC Taxi mobility analytics (12M+ records) and credit card fraud detection (1M+ transactions).
- Tech: Python, scikit-learn, PySpark, Dask.
General-purpose library for Nonparametric Combination (NPC) of permutation tests, bootstrap resampling, and multi-criteria severity ranking.
- Purpose: Pip-installable statistical toolkit for distribution-free, small-sample inference.
- Tech: Python, NumPy, SciPy.
Universal vector database migration & sync tool — migrates embeddings between Chroma, Qdrant, and other search engines.
- Purpose: One-container solution for cross-engine embedding migration.
- Tech: TypeScript, Docker, Chroma, Qdrant.
I actively contribute to major AI/ML open-source projects with bug fixes, performance improvements, and new features:
| Repository | PR | Description | Status |
|---|---|---|---|
| qdrant/qdrant | #1264 | Vector search engine improvement | ✅ Merged |
| run-llama/llama_index | #22343 | MinioReader basename collision fix | 🔍 Under Review |
| chroma-core/chroma | #7432 | Embedding search improvement | 🔍 Under Review |
| logspace-ai/langflow | #14051 | Workflow engine enhancement | 🔍 Under Review |
| lancedb/lancedb | #3661 | Retrieval pipeline fix | ✅ Approved |
| milvus-io/pymilvus | #3686 | Python SDK improvement | 🔍 Under Review |
| explodinggradients/ragas | #2850 | Evaluation framework fix | 🔍 Under Review |
| cleanlab/cleanlab | #1321 | Data-centric AI enhancement | 🔍 Under Review |
| public-apis/public-apis | #6592 | Reported 5 broken API links | 📋 Issue Filed |
C# / .NET AI Trading Agent & Portfolio Strategy Validator using SPRT, Sharpe/Sortino Ratios |
Ruby AI Financial Fraud & Anomalous Transaction Auditor with consensus verification |
Java Demographic Fairness Credit Risk Evaluator with bias metrics |
Permutation-Based Analysis of Clinical Variables in Necrotizing Fasciitis Using NPC and Bootstrap
Mathematics, MDPI (2025)
This work introduces a permutation-based, nonparametric framework for analyzing clinical variables in necrotizing fasciitis. By combining Nonparametric Combination (NPC) methodology with bootstrap techniques, the study enables robust inference under small-sample and distribution-free conditions, with an emphasis on interpretability and clinical relevance.
The study demonstrates how permutation-based inference can outperform classical parametric approaches in rare-disease clinical settings.
🔗 https://www.mdpi.com/2227-7390/13/17/2869
- permutation-based inference for small-sample biomedical studies
- interpretability under distribution shift
- robustness diagnostics for clinical ML models
- statistical foundations of explainable AI
- MCP tool development for research workflows
- 📘 math and statistics-first explanations of ML & AI
- 🧪 reproducible experiments with robust inference
- 📊 real-world clinical and analytical datasets
- 🧠 research-oriented notebooks focused on why, not just how
- 🔧 production tools — MCP servers, vector sync engines, ETL frameworks
- 📦 multi-language packages published on PyPI, NuGet, RubyGems, Maven
⭐ Thoughtful questions and rigorous discussions are always welcome.