James Brown

About Me

Data Scientist with 5+ years of experience building end-to-end machine learning pipelines, statistical models, and generative AI systems. Skilled in LLM fine-tuning, NLP, and multimodal data analysis. Backed by a strong foundation in cognitive neuroscience, with extensive experience in experimental design, behavioral data collection, and analysis using wearable technologies (e.g., HRV, EEG, eye-tracking).

Currently serving as Data Scientist at Booz Allen Hamilton, I specialize in human performance optimization, developing AI-driven solutions for warfighter readiness, and building interactive analytics platforms. I am passionate about translating complex human and machine data into actionable insights and AI-driven solutions.

Technical Skills

Python (9 yrs) R (5 yrs) SQL (3 yrs) Machine Learning (7 yrs) NLP (4 yrs) LangChain PyTorch TensorFlow Tableau Power BI Streamlit HuggingFace

Education

PhD in Biological Psychology

Virginia Tech, 2020

MS in Data Analytics & Applied Statistics

Virginia Tech, 2020

MS in Biological Psychology

Virginia Tech, 2015

BA in Psychology

Coastal Carolina University, 2011

Experience

July 2023 - Present

Data Scientist

Booz Allen Hamilton, Inc.

Designed advanced agent (Claude 3.5) powered workflow to analyze human performance data of warfighters and deliver performance optimization recommendations
Implemented machine learning model (Random Forest) to predict trainee pass/fail outcomes, enhancing early intervention strategies for the Second Air Force's 737th Training Wing
Built interactive NLP application using Hugging Face Transformers, spaCy, BERTopic, and LangChain; deployed via Streamlit for human performance optimization R&D
Developed Python-based chatbot leveraging LLMs to enable non-technical users to query and analyze war-gaming simulation outputs
Built Power BI dashboard analyzing human performance trends across Air Force training wings for command-level stakeholders
Served as Cognitive Scientist for Warrior Games Challenge 2024, integrating wearable readiness metrics and real-time dashboard analytics

July 2022 - July 2023

Senior Data Scientist

Ernst & Young, QS2

Co-developed software leveraging graph-based analytics and topic modeling to forecast emerging knowledge trends in technology domains
Built custom visualization tools for Neo4j knowledge graphs using Python and Streamlit for $50K cybersecurity initiative
Provided data engineering and BI support using BigQuery for $5M supply chain analytics project with Home Depot
Co-developed LLM-powered chatbot generating natural language summaries from Neo4j graph database
Authored generative AI thought piece to inform strategic planning and enterprise AI playbook development

July 2020 - June 2022

Senior Data Analytics Engineer

MITRE, VA Benefits

Applied NLP techniques to detect and analyze error patterns in unstructured free-text data from electronic health records
Developed hierarchical linear models investigating employee-level factors in regional claims processing performance
Engineered custom software and generated large-scale synthetic medical datasets for data privacy assessment
Co-developed PowerApps and Power BI solutions for internal MITRE and government analytics projects

February 2018 - July 2020

Research Assistant

Fralin Biomedical Research Institute

Led redesign and digitization of intake platform for Roanoke Valley Hope Initiative community health clinic
Delivered data visualizations and statistical analyses of patient data supporting engagement with regional healthcare providers

Teaching Experience

2015 - 2020

Graduate Teaching Assistant

Virginia Tech, Department of Psychology

Instructed undergraduate psychology laboratory courses
Facilitated hands-on learning experiences in experimental psychology
Mentored students in research methodology and statistical analysis
Developed and delivered course content for Introduction to Psychology
Supervised undergraduate research assistants in cognitive neuroscience experiments

Summer 2017

Research Mentor & Instructor

NIH/NIMH Learning and Plasticity Section

Mentored undergraduate students in neuroimaging research methods
Taught fMRI data analysis techniques and experimental design
Supervised student research projects in visual neuroscience
Collaborated with Dr. Chris Baker and Dr. Elijah Merriam on educational initiatives

2011 - 2015

Undergraduate Research Coordinator

Life Skills Program

Coordinated educational workshops on psychological research methods
Supervised undergraduate peer mentoring programs
Developed curriculum for research methodology training

Projects

Warfighter Performance Optimization

Advanced agent-powered workflow using Claude 3.5 to analyze human performance data and deliver mission-critical optimization recommendations

Claude 3.5 AI Agents Human Performance

Trainee Success Prediction Model

Random Forest model predicting pass/fail outcomes for Air Force trainees, enabling early intervention strategies

Python Random Forest Predictive Modeling

NLP Text Analysis Platform

Interactive application using Hugging Face, spaCy, and BERTopic for topic extraction, sentiment analysis, and text summarization

HuggingFace BERTopic Streamlit LangChain

War-Gaming Simulation Chatbot

LLM-powered chatbot enabling non-technical users to query and analyze war-gaming simulation outputs

Python LLMs OpenAI Anthropic

Neo4j Knowledge Graph Visualizer

Custom visualization tools for knowledge graphs supporting $50K cybersecurity initiative

Neo4j Python Streamlit Graph Analytics

Air Force Performance Dashboard

Power BI dashboard analyzing human performance trends across training wings for command-level decision support

Power BI Data Visualization Analytics

EHR Error Pattern Detection

NLP system for detecting and analyzing error patterns in unstructured electronic health records data

NLP Python Healthcare

Warrior Games Analytics Platform

Real-time analytics and biometric monitoring system for athletes using wearable readiness and recovery metrics

HRV Biometrics Real-time Analytics

Publications

Predictive Modeling of Cognitive Performance Using Multimodal Data

Journal of Cognitive Neuroscience, 2024

A comprehensive study on using machine learning to predict cognitive performance from combined EEG and behavioral data.

Statistical Approaches to Understanding Neural Connectivity Patterns

NeuroImage, 2023

Novel statistical methods for analyzing functional connectivity in neuroimaging data.

Data-Driven Insights into Stress Response Mechanisms

Biological Psychology, 2023

Large-scale analysis of physiological stress responses using advanced analytics techniques.