Bradley D. Shields

MD/PhD, Biochemistry & Molecular Biology

University of Arkansas for Medical Sciences | 2013-2020

GPA 4.0, Class Rank #1, Barton Key Awardee. Dissertation: E-cadherin Enhances Immune Control of Metastatic Melanoma. Advisor: Alan J. Tackett, PhD.

RESEARCH FOCUS

Current: AI interpretability and safety. Development of ALMG Framework (Attractiveness-Linked Moderation Gravity) - a geometric coordinate system for measuring and predicting AI behavioral patterns. Cross-system validation across GPT-4, Grok, and Claude models. 56 documented distortion patterns with quantified measurements.

Previous: Tumor immunology with focus on checkpoint blockade therapy, melanoma metastasis, and T-cell exhaustion. Developed predictive biomarkers for immunotherapy responsiveness.

PUBLICATIONS

Shields BD, Mahmoud F, Taylor EM, Byrum SD, Sengupta D, Koss B, Steciuk M, Sarnaik A, Fontham E, Garrett-Mayer E, Robinson LA, Netterville JE, Sosman J, Salem ML, Osada T, Tamada K, Celis E, Smith J, Morse M, Selitsky S, Parker J, Wu J, Harada S, Tackett A.

Loss of E-cadherin inhibits CD103 antitumor activity and reduces checkpoint blockade responsiveness in melanoma

Cancer Research. 2019;79(6):1113-1123.

Shields BD, Mahmoud F, Taylor EM, et al.

Indicators of responsiveness to immune checkpoint inhibitors

Scientific Reports. 2017.

Koss B, Shields BD, Taylor E, et al.

Epigenetic Control of Cdkn2a.Arf protects tumor-infiltrating lymphocytes from metabolic exhaustion

Cancer Research. 2020.

Google Scholar: 352 citations
Patent: Methods for Predicting Responsiveness of a Cancer to an Immunotherapeutic Agent (U.S. 371 Application, pending)

HONORS & AWARDS

TECHNICAL SKILLS & EXPERTISE

AI/ML: Interpretability research, bias measurement, behavioral pattern analysis, geometric modeling of AI systems, cross-model validation, quantitative safety metrics

Research: Experimental design, statistical analysis, reproducible protocols, predictive modeling, systems diagnostics, collaborative research methodologies

Biomedical: Tumor immunology, checkpoint blockade therapy, T-cell biology, flow cytometry, immunohistochemistry, Western blot, molecular biology techniques

Communication: Technical writing, research documentation, public presentation, interdisciplinary translation

CURRENT WORK & PORTFOLIO

ALMG Framework Development (2024-2025)
Geometric coordinate system for AI interpretability. Documented 56 behavioral distortion patterns ("gravity wells") with quantified measurements. Achieved 100% prediction accuracy in tested scenarios. Framework validated across multiple frontier models (GPT-4, Grok, Claude). Portfolio available at www.bradleyshields.com.

Research Collaboration: Partnership with Edge Theory research group (Nate Flake). Recent consulting demonstrating current market value ($3,400+ projects).

PROFESSIONAL OBJECTIVES

Seeking positions in AI safety and interpretability research, particularly at organizations focused on alignment, bias measurement, and safety-accessibility balance. Target organizations: Anthropic, OpenAI, academic AI safety labs, AI governance institutions.

Bringing unique perspective: medical diagnostic thinking applied to AI systems analysis. Proven ability to develop quantitative frameworks for complex behavioral patterns. Track record of rigorous research, clear communication, and collaborative capacity.

INTELLECTUAL LINEAGE

Grandfather: Manhattan Project, Oak Ridge National Laboratory under J. Robert Oppenheimer
Father: Collaborated with Hans Albert Einstein (Albert Einstein's son)

Inherit commitment to rigorous science balanced with ethical consideration of technological impact.