CV

Education

University of Central Florida — M.S. in Statistics and Data Science (Fall 2026 – Expected May 2028)

• Graduate Teaching Assistant (20 hrs/week)
• Full tuition remission + health insurance

University of Central Florida, Burnett Honors College — B.S. in Data Science, Minor in Statistics (May 2026)

• GPA: 3.83
• Burnett Honors College Scholar

Undergraduate Research

AI Undergraduate Researcher, UCF (Spring 2026 – Present)

• Active research project — details coming soon

Honors Undergraduate Thesis (Summer – Fall 2025)

• “Theoretical Analysis of CNNs for Automatic Seizure Detection in EEG Signals”
• First Data Science undergraduate to publish in UCF STARS Honors Undergraduate Thesis repository
• Published: stars.library.ucf.edu/hut2024/462
• Advised by Dr. Chudamani Poudyal (SDMSS, UCF)
• Results: 97% accuracy, 0.99 AUC, Lipschitz stability bounds (L = 24.72)
• Poster presentations: Burnett Honors College Family Weekend (Sep 2025), Student Scholar Symposium (Mar 2026)

Industry Experience

AI Associate Developer, Insurity Inc. (Fall 2025 – Present, Remote/Part-Time)

• Built and deployed RAG/LLM prototypes using Anthropic, OpenAI, and Replit APIs for core insurance product features; presented POCs to senior engineers
• Migrated 175 SSRS reports to Power BI in 3 weeks vs. 3–4 month industry projection — $700,000 cost avoidance, 17x industry speed, processing 1.1M+ lines XML across 187 DB connections with 100% pass rate
• Architected dual-layer HIPAA audit system covering 23 PHI tables (9,196 lines C# + SQL) with dynamic SQL triggers achieving 85–96% storage reduction and zero audit gaps; implemented 3-tier RBAC + field masking
• Delivered enterprise SDK integration (10,723 LOC, 334 automated test checks) using provider-pattern architecture + OAuth2 lifecycle management in 22 days vs. 3–6 month industry standard

Research Assistant (Data Science) Intern, Design Interactive Inc. (May – Aug 2024, Part-Time)

• Built automated data validation pipelines in Python + SQL for large human-systems research datasets
• Led QA on VR/AR Unity simulations; deployed builds to hardware; validated eye-tracking software for participant gaze capture
• Conducted statistical analysis on behavioral data; led pilot studies (~300 participants)

Projects

DataSci-Coder (Spring 2026)

• Fine-tuned Qwen2.5-Coder-14B via QLoRA on 10,795 curated data science examples spanning statistics, ML, and deep learning
• Training: L40S GPU, 1.9 hours, QLoRA r=16 alpha=32, 4-bit quantization
• Results: 93.3% instruction compliance vs 91.4% base (+1.9%), 10/10 code-only output vs 6/10 base, 100% code ratio vs 87.9% base (+12.1%)

• HuggingFace: jsmall12/DataSci-Coder-14B-LoRA

  GitHub: jacksonSmall/DataSci-Coder

Virtual Patient UCF × ITESM (Spring 2026)

• Capstone project (UCF + Tecnologico de Monterrey, team of 6) testing whether video-structured AI context improves automated medical communication training
• 3 experiments: A/B testing 40 OSCE transcripts, 6-dimension rubric scorer (emotional attunement, naturalness, nonverbal responsiveness, phase appropriateness, medical relevance, response quality)
• Benchmarked Llama 3.1 8B / BioMistral 7B / Qwen 2.5 7B; median 5.9s response latency, instant feedback

Neuro-Wave Analytics (Fall 2025)

• Bayesian Physics-Informed Graph Neural Network for seizure propagation modeling across EEG electrode networks
• Uncertainty quantification; PyTorch Geometric; extension of honors thesis work

AI Driver Risk Scoring Platform (Fall 2025)

• End-to-end driver risk assessment from telematics data; 47 engineered behavioral risk features
• ROC-AUC 0.98; real-time Streamlit dashboard with interactive Plotly maps, SHAP feature importance, gamification rewards

Skills

Languages: Python, R, SQL, Julia, Bash, JavaScript

ML & Deep Learning: PyTorch, TensorFlow, Scikit-learn, XGBoost, PyTorch Geometric, QLoRA/PEFT, Hugging Face Transformers, SMOTE

Data Science: Pandas, NumPy, SciPy, Statsmodels, Matplotlib, Seaborn, Plotly, SHAP

LLM & NLP: RAG systems, LLM fine-tuning, prompt engineering, Anthropic/OpenAI/Azure APIs, open-source LLMs (Llama, Qwen, BioMistral)

Statistics & Methods: Time Series, Survival Analysis, Stochastic Processes, Bayesian Methods, Multivariate Analysis, Signal Processing, A/B Testing, Cross-Validation

Tools & Cloud: Azure, AWS, Docker, Git/GitHub, FastAPI, Streamlit, Power BI, SLURM/HPC

Leadership & Certifications

Campus & Community

• CRU at UCF — Youth Group Leader & Worship Musician (Fall 2022 – Spring 2026)
• Pegasus Math Club — STEM Day presenter (2023)
• AMS Spring Southeastern Sectional Meeting, FSU (March 2024)

Honors & Recognition

• Burnett Honors College Scholar, UCF
• BSA Eagle Scout (2020 – Present)
• Epilepsy Foundation Ambassador (2019 – Present)

Certifications

• Microsoft Azure AI Essentials (December 2025)
• Anaconda Python for Data Science (December 2025)

Publications

Small, J. T. (2025). Theoretical Analysis of CNNs for Automatic Seizure Detection in EEG Signals. UCF STARS Honors Undergraduate Theses, No. 462. https://stars.library.ucf.edu/hut2024/462