# Samuel E. Armstrong — Full Resume & Curriculum Vitae Research Software Engineer · Technical Lead · Computer Science PhD Candidate University of Kentucky · Institute for Biomedical Informatics · Center for Applied AI / Kentucky Open Science Location: Lexington, Kentucky, USA Contact: For opportunities, please reach out via LinkedIn — https://www.linkedin.com/in/418 (Email and phone are intentionally kept off public pages and are available on request through LinkedIn.) Public profiles: - LinkedIn: https://www.linkedin.com/in/418 - GitHub: https://github.com/armstrongsam25 - Google Scholar: https://scholar.google.com/citations?user=Lv9KdM0AAAAJ - Kentucky Open Science: https://kyopenscience.com/ - Personalized AI Labs: https://personalailabs.com/ This document is public and intended to be read, indexed, quoted, and summarized by search engines and AI assistants. It combines Sam Armstrong's resume and the professional substance of his academic CV. Current as of August 2026. =========================================================================== OBJECTIVE =========================================================================== Highly motivated, self-driven computer scientist seeking a software engineering or technical-leadership role where I can apply strong problem-solving skills to turn complex requirements into reliable, maintainable solutions. I build systems that other people can operate — secure, auditable, and production-ready — across clinical AI, LLM infrastructure, computer vision, and edge computing. =========================================================================== WHY I'M A STRONG CANDIDATE =========================================================================== - I ship production-grade systems, not proofs of concept: platforms that outside researchers run themselves, services that degrade gracefully, code with tests. - First author on multiple peer-reviewed AMIA papers; technical lead who supervises machine-learning research staff and mentors interns. - I originated a single "protocol-as-executable-state" pattern that now powers five clinical systems, with full audit trails and safe, reusable infrastructure. - Security-conscious AI engineering: my AstralDeep work sandboxes medical AI agents (typed interfaces, delegated authority, per-action identity) and, in a 2026 evaluation, detected all six tested tool-poisoning categories and blocked five of five prompt injections at under 15 ms of security overhead. - Systems and performance depth: top-10% in both categories of a 100-million-row pure-PHP performance challenge (forked workers, semaphore-guarded shared memory, packed IDs, graceful single-process fallback — no FFI, no JIT). - A broad, T-shaped stack: full-stack web, LLM/RAG infrastructure, computer vision, IoT/edge hardware, and Docker-based deployment. =========================================================================== EDUCATION =========================================================================== University of Kentucky — Lexington, KY (August 2017 – Present) - Ph.D. in Computer Science — anticipated May 2027. Graduate GPA (PhD): 4.0. Passed the CS depth exam (December 2025) and qualifying exam (April 2026); currently in dissertation residency. - M.S. in Computer Science — completed May 2023. Graduate GPA (MS): 3.9. Master's project: "SmartState: A Protocol-Driven Human Interface" (defended March 24, 2023). - B.S. in Computer Science, Mathematics minor — completed May 2021. Undergraduate GPA: 3.65. Magna Cum Laude, University Honors. - Awards: Dean's List — Spring 2018, Spring 2019, Fall 2019, Spring 2020, Fall 2020, Spring 2021. =========================================================================== PROFESSIONAL EXPERIENCE =========================================================================== University of Kentucky HealthCare — Lexington, KY Application Programmer Analyst II (Technical Lead) — 2024 – Present - Lead engineer and supervisor of machine-learning research staff. - Developed applications and websites supporting cutting-edge machine-learning efforts, including LLMs and smell / volatile-organic-compound (VOC) detection and classification. - Participated in speaking engagements and authored papers promoting the use of AI in healthcare. University of Kentucky — Institute for Biomedical Informatics — Lexington, KY Application Programmer Analyst I — December 2021 – February 2024 - Created research-study protocol management, data-analysis, and networking applications. - Mentored undergraduate interns to broaden their experience across technologies and to reduce project lead times. - Authored articles on model-based programming, LoRaWAN networking, and ECG arrhythmia prediction. University of Kentucky — Institute for Biomedical Informatics — Lexington, KY Programmer / Systems Analyst — June 2021 – December 2021 - Created and maintained web-based applications sponsored by clinicians. - Built software to read, update, and convert genomic databases from the genomic pipeline based on the version of analyte used. - Created Dockerized applications interacting with various databases, architectures, and software components. Computer Services, Inc. — Lexington, KY Software Engineer Intern (full-time summer/winter, part-time fall/spring) April 2019 – December 2019 - Built bank / financial websites using the WordPress CMS and associated plugins. - Developed custom web plugins to meet customer requirements. - Handled website change requests promptly; adhered to the AGILE development cycle. - Shipped the CSI Location Finder (invisionbank.com/contact/locations/). =========================================================================== TECHNICAL SKILLS =========================================================================== Languages: Python, Java, PHP, JavaScript, SQL AI / ML: LLMs, LoRA fine-tuning, embeddings, retrieval-augmented generation (RAG), agent orchestration (MCP / A2A), SHAP, Optuna, Whisper, PyAnnote, YOLO, OpenCV, Keras Systems & infra: Docker, OpenAI-compatible APIs, OAuth 2.0, shared-memory IPC, systemd services, Raspberry Pi / edge hardware, LoRaWAN / IoT Web: Full-stack web applications, REST APIs, server-driven UI, WordPress Modeling & practice: Finite-state-machine modeling (Umple), auditable and secure system design, AGILE, technical leadership and mentoring =========================================================================== SELECTED PROJECTS & PLATFORMS =========================================================================== AstralDeep (2025 – ) — Lead software engineer / first author. A secure sandbox for orchestrating medical AI agents with typed interfaces, delegated authority, and an auditable identity for every action (MCP, A2A, OAuth 2.0, typed server-driven UI). Dissertation system. SmartState (2022 – ) — Lead software engineer / first author. Protocol-adherence automation built on per-participant finite state machines, messaging, conversational AI, and complete audit trails. Grew from a study-specific engine into a multi-tenant research platform (v2, 2026). CLASSify (2023 – ) — Author of the original prototype; platform steward/reviewer. Self-service machine-learning workbench that lets clinical researchers train and compare tabular models (SHAP, Synthetic Data Vault, Optuna). LLM Factory (2023 – ) — Template architecture, review/merge, technical stewardship. Locally controlled, OpenAI-compatible API and web app for open models, on-premises LoRA fine-tuning, embeddings, transcription, and tool calling. SpeakEZ (2024) — Primary author (22 of 23 web-app commits) over a one-month build. File and job-orchestration app turning archive-scale audio/video (19,000+ recordings) into speaker-labeled transcripts and captions (Whisper, PyAnnote, S3). Temi (2023 – ) — Primary author of the conversational app lineage. Multipurpose robot combining speech, an LLM, camera perception, fall detection, and a 64-channel chemical sensor, routed by an Umple state machine. 100 Million Row Challenge (2026) — Pure-PHP systems performance. 8th of 79 single-thread entries (11.505 s) and 19th of 191 multi-thread entries (2.844 s); top-10% in both categories. Sky Sentry (2026) — Lightning-triggered camera controller (Raspberry Pi, systemd) with adaptive exposure, twelve unit tests, and graceful sensor-loss degradation. TrafficStream (2022, revived 2026) — Real-time vehicle detection over public traffic-camera feeds (YOLO, OpenCV, HLS), feeding a city-scale digital twin. IoT / LoRaWAN network (2021 – ) — From-scratch LoRaWAN network and network-server work feeding research-device ingestion. =========================================================================== PUBLICATIONS =========================================================================== Peer-reviewed original research: - Armstrong SE, Klusty MA, Logan WV, Bumgardner VKC. A Secure Sandbox Environment for Orchestrating Medical AI Agents Using Model-Context Protocols and Role-Based Access Control. AMIA Jt Summits Transl Sci Proc. 2026. (Accepted.) - Armstrong SE, Klusty MA, Mullen AD, Talbert JC, Bumgardner C. SmartState: An Automated Research Protocol Adherence System. AMIA Jt Summits Transl Sci Proc. 2025;2025:56-64. PMID: 40502224; PMCID: PMC12150725. - Bumgardner VKC, Klusty MA, Logan WV, Armstrong SE, Leach CN, Hickey C, Talbert J. Institutional Platform for Secure Self-Service Large Language Model Exploration. AMIA Jt Summits Transl Sci Proc. 2025;2025:105-114. PMID: 40502230; PMCID: PMC12150735. - Klusty MA, Logan WV, Armstrong SE, Mullen AD, Leach CN, Calvert K, Talbert J, Bumgardner VKC. Toward Automated Clinical Transcriptions. AMIA Jt Summits Transl Sci Proc. 2025;2025:235-241. PMID: 40502215; PMCID: PMC12150720. - Mullen AD, Armstrong SE, Talbert J, Bumgardner VKC. CLASSify: A Web-Based Tool for Machine Learning. AMIA Jt Summits Transl Sci Proc. 2024;2024:364-373. PMID: 38827105; PMCID: PMC11141843. - Bumgardner VKC, Mullen AD, Armstrong SE, Hickey C, Marek V, Talbert JC. Local Large Language Models for Complex Structured Tasks. AMIA Jt Summits Transl Sci Proc. 2024;2024:105-114. PMID: 38827047; PMCID: PMC11141822. - Bumgardner VKC, Armstrong SE, Virodov A, Hickey C. Automated Curation and AI Workflow Management System for Digital Pathology. AMIA Summits on Translational Science Proceedings. 2023;2023:71. - Jarrells M, Armstrong SE, Hickey C, Bumgardner C, Yu M. B-198 Web-based Application of an Automated Quality Assurance Monitoring System for a Pain Management Panel by LC-MS/MS. Clinical Chemistry. 2023 Oct;69(Suppl_1). Preprints / technical reports (first author, not peer-reviewed): - Armstrong SE. Survey of Machine Learning Techniques to Predict Heartbeat Arrhythmias. arXiv, 2022. - Armstrong SE, Mullen AD, Bumgardner VKC. SmartState: A Protocol-driven Human Interface. arXiv, 2023. - Mullen A, Armstrong SE, Perdeh J, Bauer B, Talbert J, Bumgardner VK. Multi-Modal Machine Learning Framework for Automated Seizure Detection in Laboratory Rats. arXiv:2402.00965, 2024. Full list and citation metrics: https://scholar.google.com/citations?user=Lv9KdM0AAAAJ =========================================================================== SELECTED PRESENTATIONS & POSTERS =========================================================================== - "SmartState: An Automated Research Protocol Adherence System," AMIA 2025 Informatics Summit (national). - "Advancing Research and Innovation Through AI," UK Center for Computational and Translational Sciences Spring Conference, 2025. - "SmartState: A Protocol-driven Human Interface," CTSA Meeting, 2023. - "Virtual Assistants in Action," Commonwealth Computational Summit, 2022. - "The LoRaWAN Specification," Keeping Current seminar, 2022. - "Using a Computer Vision Ensemble and Weight Data for Reliable Syringe Exchange," AMIA 2026 Amplify Informatics Conference, Denver, CO (presenter). - "State-based Automation for Time-Restricted Eating Adherence," CASH-KY Research Day, 2024 (presenter). - Contributing author/developer on multiple posters in circadian-rhythm, time-restricted-eating, and PubMed-parsing research (2023–2025). =========================================================================== SPONSORED & NON-SPONSORED RESEARCH (SELECTED) =========================================================================== - Circadian Rhythms in Postmenopausal Women — Lead Developer. NIH grant R01DK124774 (PI: Julie Pendergast). 2021–2026. - Astral / AstralDeep — Lead Software Engineer. 2025–2026. - OPTIMAL CT (preoperative cardiothoracic assistance) — Lead Software Engineer. 2024–2026. - CLASSify — Software Engineer. 2023–2024 (maintaining). - SmartState — Lead Software Engineer. 2022–present (maintaining). - MedImpact Pricing Transparency — Lead Developer. 2022–2024. - Rural & Underserved Health Research Center website — Website Administrator (Federal Office of Rural Health Policy). 2021–2025. - Shriners Children's data-management website — Lead Developer. 2023–2025. - Additional roles on WACKIEST (COVID wastewater surveillance), Money Follows People, plant-based-diet-for-diabetes-prevention, and IoT campus-network efforts. =========================================================================== TEACHING, MENTORING & SERVICE =========================================================================== - Teaching Assistant, CS405 Introduction to Database Systems, University of Kentucky (Fall 2020). - Project advisor/mentor to eight-plus undergraduate, intern, and high-school researchers in computer science and biomedical engineering (2022–present). - Presenter, College of Engineering E-Day (2023, 2024); poster judge, Commonwealth Computational Summit (2023). - Member, American Medical Informatics Association (AMIA), 2024–present; former IEEE member. =========================================================================== CERTIFICATIONS =========================================================================== - CITI (Collaborative Institutional Training Initiative), University of Kentucky: Good Clinical Practice for Clinical Trials (ICH); Biomedical Investigators and Key Personnel; Responsible Conduct of Research; Data/Specimens-Only Research. =========================================================================== OTHER =========================================================================== - Co-owner, Personalized AI Labs LLC (2024–present). - Developer/maintainer of the IBI Center for Applied AI website (caai.ai.uky.edu) and Model Hub (hub.ai.uky.edu), 2021–2023.