Career Objective

AI expert with 10+ years of experience in LLMs, knowledge graphs, and healthcare IT, specializing in developing scalable AI solutions. Proven track record in AI research, compliance, and real-world applications. Published 9 papers (76 citations) and contributed to ONR, NSF, and PPN-funded projects. Passionate about transforming AI innovations into impactful, enterprise-ready solutions that drive efficiency and business growth.

Key Competencies

Large Language Model (LLM) Natural Language Processing Knowledge Graphs
Data Processing Data Analysis Data Visualization
Hyperparameter Optimization Software Engineering Healthcare IT

Experience

Research Postdoctoral Fellow, Children’s National Hospital — Washington, D.C.
Dec 2023 – Present

  • Developed mhGPT, a 1.98B parameter transformer trained on 11M social media posts & 50K PubMed articles, surpassing Google Gemma-2B and MentaLLaMA in classification and NER accuracy.
  • Building multimodal deep-learning models using TriNetX data for early pediatric mental health detection, enhancing resilience through predictive interventions.

Knowledge Engineering Intern, Bosch USA — Sunnyvale, CA
May 2023 – August 2023

  • Designed and deployed the Manufacturing Event Knowledge Graph (MAKE), converting factory logs into actionable insights for predictive maintenance & revenue optimization.

Research Assistant, University of Maryland, Baltimore County — Baltimore, MD
June 2018 – April 2023

  • Developed a federated healthcare data exchange and compliance framework, automating secure health data interoperability using knowledge graphs (CDC COVID-19, HIPAA, Synthea RDF).
  • Built a clinical decision-support system analyzing ACLR patients’ movement, securing $1M NIH funding for wearable sensor research in breast cancer rehabilitation.
  • Refactored the Data Quality Toolkit (DQT) database, reducing user time by 40% in detecting & cleaning health data defects.

Projects

Synthea RDF
github.com/leroykim/synthea-rdf

  • Developed a Semantic Web representation of synthetic patient data and a CSV-to-RDF conversion tool for inference across patient data modalities.
  • Work adopted as the standard structured data model at The Hilltop Institute, UMBC, and later extended with blockchain for secure claim data validation.

Skills

  • AI & Machine Learning: Python, Hugging Face Transformers, DeepSpeed, Ray Tune, PyTorch, NumPy, Pandas, Spacy, scikit-learn, Matplotlib, Plotly, Flyte, Dask
  • Knowledge Graphs & Data Engineering: Apache Jena Fuseki, SPARQL, Protégé, Rdflib, Turtle, OWL, NetworkX, MATLAB, R, SQL, SQLite, Java, Spring, Flask
  • Healthcare Data Standardization & Compliance: HL7 (FHIR, CDA), HIPAA Knowledge Graph
  • Cloud & DevOps: AWS EC2, Docker, Slurm, Airflow, GitHub, Fossil SCM, Jupyter Notebook, Emacs

Education

  • Ph.D. in Information Systems, University of Maryland - Baltimore County (Dec 2023)
  • M.S. in Information Systems, University of Maryland - Baltimore County (June 2020)
  • B.S. in Computer Science and Engineering, Kyungpook National University (June 2018)