Hi, I'm Glenn Dalbey
AI Engineer & Data Scientist specializing in Data Analytics, multi-modal systems, and production ML deployments
About Glenn
Passionate AI engineer with a proven track record of building and deploying production machine learning systems that make a real-world impact.
My Journey
My journey into AI and data science began with a fascination for solving complex problems through technology. What started as curiosity about machine learning algorithms evolved into a passion for building systems that can genuinely improve people's lives.
The turning point came when I developed Apollo Healthcare Connect - a multi-modal AI system that combines natural language processing with computer vision to provide medical triage with 93.8% accuracy. Seeing this system develop into a tool that could help real patients find appropriate care reinforced my commitment to practical, impactful AI applications.
Today, I specialize in bridging the gap between cutting-edge AI research and real-world deployment, with particular expertise in data analytics, multi-modal systems, and production ML pipelines.
Education
Advanced graduate program focused on machine learning, statistical analysis, and practical AI applications. Completed advanced capstone projects in healthcare AI and sports analytics with production deployment.
Key Achievements
By the Numbers
Featured Projects
Explore my portfolio of AI and data science projects, from healthcare applications to sports analytics and optimization solutions.
NFL_Big_Data_Bowl_2026
Kaggle Bronze Medal (Top 8% of 1,134 teams) - Deep learning solution for predicting NFL player trajectories from tracking data. Explored 15+ architectures across 847+ experiments with systematic hyperparameter optimization.
missing-persons-outlier-detection
Multi-method statistical and ML pipeline analyzing 41,200 NamUs cases across 55 jurisdictions and 101 years. Applies 7 outlier detection methods, Isolation Forest and LOF ensemble, spatial autocorrelation (Moran's I), and ARIMA forecasting. Identified the I-35 trafficking corridor (+170% acceleration) and validated against known serial killers. Live Streamlit dashboard.
Apollo_Healthcare_Connect
Production-deployed multi-modal AI healthcare triage system achieving 93.8% combined accuracy. Analyzes text symptoms (DistilBERT, 94%) and medical images (5-model CNN ensemble, 98% burn classification) across 8,085 images with 29.7:1 class imbalance handling. WGU MS Data Science Capstone. Live at apollohealthcareconnect.com.
RSNA_Intracranial_Aneurysm_Detection
3D deep learning solution for detecting intracranial aneurysms from CT angiography. Trained 105 models (21 architectures x 5 folds), tested 51 ensemble configurations achieving best AUC 0.8624. Key finding: smaller models significantly outperform larger ones on limited medical data (r=-0.42, p<0.01).
OceanEternaIn Progress
Multi-month engineering effort building a high-performance local RAG system. Originally prototyped in Python but rewritten from scratch in C++17 for performance — evolved through 4 major versions with systematic optimization. Search engine core indexes 2.45 billion tokens across 5M+ chunks with 0-42ms search latency (avg 12ms, down from 500ms in v1) and 12-second cold startup (down from 41s). Runs entirely on CPU with minimal RAM. Dual LZ4/Zstd compression with auto-format detection. Conversations and queries continuously indexed with intelligent tagging; supports ingesting any file type to grow the corpus. 47 tests at near 100% accuracy, 15 REST API endpoints, zero per-query costs — no GPU, Docker, database, or cloud required. Next: LLM chat interface and MCP tool integration for terminal-based AI workflows and project knowledge management.
OE-OSIn Progress
Distributed AI orchestration platform for a private multi-node GPU cluster. Features three-tier LLM routing (local Ollama to cheap API to Claude Opus) reducing costs by routing ~80% of requests to free local models, triple-layer RAG memory (BM25 over 5M+ chunks, ChromaDB semantic search, Redis session cache), 18 MCP-compatible tools, and a multi-agent sandbox where 4 LLM personas deliberate at zero API cost. 4,200+ lines of async Python on FastAPI.
opportunity-intelligence
AI-powered market analysis assistant for senior living opportunity evaluation. Agentic 3-call LLM pipeline that pulls public data from Census Bureau and CMS Care Compare, runs it through a reusable analytics library of 22 statistical methods, then uses an LLM to direct the analysis and synthesize findings into an executive briefing with full citations. 14 analyses executed across two LLM-directed passes, 622-word briefing with source tags on every claim, total cost 6 cents.
Blue-Zones-Longevity-Analysis
Longitudinal statistical analysis of life expectancy trends across Blue Zone countries vs. 88-nation global baseline, 1960-2023. Formal hypothesis testing with bootstrap confidence intervals, GDP-controlled partial correlations, sigma and beta convergence testing, and COVID impact analysis. Live Streamlit dashboard.
business-analytics-AI-platform
AI-powered Excel analytics platform for Thompson Parking & Mobility Consultants. Upload Excel files, get instant business insights, generate professional charts, and chat with your data using natural language queries.
Technical Skills
A comprehensive overview of my technical expertise across AI/ML, data science, and software development technologies.
Programming Languages9
AI/ML Frameworks11
Web Frameworks5
Cloud & Deployment4
Data & Analytics7
AI Specializations12
Resume & CV
Complete professional resume showcasing my experience in AI, data science, and machine learning.
Glenn Dalbey
Data Science & Analytics Professional
Professional Summary
Data Scientist with an MS in Data Science who builds things that actually work. I enjoy building and training models on my own multi-GPU homelab, deploying new techniques, and discovering unique ways to solve problems. Whether it's analysis, research, or projections, I dig into how something ticks, figure out why, and extract information that drives better outcomes. My multi-node homelab stack lets me run production applications, train models, and run MLflow, all on my own hardware. 20+ projects on GitHub covering deep learning, medical imaging, NLP, computer vision, and analytics.
Core Technical Skills
Programming & Deep Learning
Neural Network Architectures
ML & Infrastructure
Specializations
Selected Projects
NFL Big Data Bowl 2026 - Kaggle Bronze Medal
GitHubPlayer Trajectory Prediction | Bronze Medal - Top 8% of 1,134 teams
- • Bronze Medal in prestigious Kaggle competition predicting NFL player trajectories
- • 847+ experiments across 15+ architectures (ST Transformers, GRU, CNN, Perceiver IO)
- • Engineered 167 features with Voronoi tessellation and geometric attention
RSNA Intracranial Aneurysm Detection
3D Medical Imaging | 105 Models Trained
- • Trained 105 deep learning models (21 architectures × 5 folds) for CT angiography
- • Best ensemble AUC 0.8624; discovered smaller models outperform larger on limited data
- • Complete pipeline: DICOM→NIfTI→ROI→Training→Ensemble on 4 GPUs
Apollo Healthcare Connect
Live SystemProduction Multi-modal AI Healthcare Triage | MS Capstone
- • Live production system achieving 93.8% accuracy with sub-second response
- • 5-model ensemble combining DistilBERT and CNNs; handled 29.7:1 class imbalance
Education
Master of Science in Data Science
Western Governors University
Capstone: Multi-modal AI healthcare triage system (production deployed)
Bachelor of Science in Data Analytics
Western Governors University
Capstone: NFL Rookie Wide Receiver Performance Prediction Model
Key Accomplishments
- Kaggle Bronze Medal - NFL Big Data Bowl 2026 (Top 8% of 1,134 teams)
- 847+ deep learning experiments across 15+ neural network architectures
- Trained 105 3D medical imaging models achieving AUC 0.8624 ensemble
- Production healthcare AI achieving 93.8% accuracy with sub-second response
- Analyzed 41,200 cases identifying trafficking patterns at up to 46.86σ significance
- Published 20+ open-source projects on GitHub
Get In Touch
Ready to collaborate on your next AI project? Let's discuss how my expertise in machine learning and data science can help bring your ideas to life.
Let's Connect
I'm always interested in discussing new opportunities, whether you're looking for an AI consultant, data scientist, or full-time team member. Feel free to reach out for:
- AI & Machine Learning Consulting
- Healthcare AI Solutions
- Data Science Projects
- Full-time Opportunities
- Technical Collaboration
I typically respond to all inquiries within 24 hours.