Hi, my name is
Harrison Smith.
I'm an
Bridging LLM orchestration, RAG architecture, and production-grade business intelligence. Pursuing an M.S. in Applied AI at Neumont — building systems that work in the real world.
01. About Me
I'm an AI Engineer and BI Analyst with a track record of turning complex data infrastructure and AI systems into real business outcomes. I hold a B.S. in Computer Science from Weber State University and am completing my M.S. in Applied Artificial Intelligence at Neumont University (expected Sept 2026), with hands-on focus on ETL pipelines, data sanitization for model training, and Retrieval-Augmented Generation architectures.
At BlytzPay, I serve as the technical bridge between stakeholders and data infrastructure — architecting BigQuery pipelines, engineering Tableau dashboards, and supporting human-in-the-loop evaluation of internal AI agents. I've published a desktop tool to the Apple App Store, built agentic systems orchestrating multiple LLMs through RAG memory layers, and led intern teams on full-stack products that influenced state-level policy data.
My background includes enterprise roles at Hill Air Force Base, Instructure (Canvas), Microsoft, and Control4 — a foundation that shapes how I approach AI systems: with the discipline, reliability, and attention to production concerns that those environments demand.
Technologies I work with:
- ▹Python & JavaScript
- ▹TensorFlow / PyTorch
- ▹LLM Orchestration & RAG
- ▹Google BigQuery
- ▹Tableau / PowerBI
- ▹SQL & ETL Pipelines
02. Experience
Business Intelligence Analyst @ BlytzPay
Nov 2025 – Present
- ▹Serve as the technical bridge between internal stakeholders and data infrastructure, translating business needs into actionable Google BigQuery data pulls and Tableau dashboards.
- ▹Support continuous human-in-the-loop training and evaluation of internal AI agents, delivering targeted prompt refinement to resolve tool-execution errors and improve response quality.
Software & IT Services @ Common Thread
Mar 2024 – Present
- ▹Execute backend troubleshooting, ongoing software development, and SQL database management to support a variety of IT operations and client requirements.
Data Analyst Coordinator & AI Trainer @ Salt Lake Community College
May 2023 – Nov 2025
- ▹Instructed specialized courses in AI, Robotics, and 3D Printing for underserved youth through the Verizon Innovative Learning program.
- ▹Co-developed and taught "AI @ Work" — an innovative curriculum prototype focused on prompt engineering and practical AI workplace integration.
- ▹Engineered data visualizations using SQL and JavaScript to align regional training programs with local career opportunity data.
- ▹Built and published the Canvas SIS Prep Tool to the Apple App Store — a cross-platform desktop utility adopted by the Workforce Training team.
Independent Technical Consultant & AI Developer @ PROG Foundation · Acord.AI
Ongoing
- ▹Deliver technical training, step-in instruction, and curriculum facilitation to empower diverse audiences in integrating AI workflows and emerging technologies.
Software Development & Systems Roles @ Hill AFB · Instructure · Microsoft · Control4
2015 – 2023
- ▹Held comprehensive software development, data analysis, and system support positions across defense, EdTech, and enterprise technology sectors.
- ▹Built a foundation in production-grade engineering discipline — large codebases, cross-functional collaboration, and zero-downtime deployments.
03. Skills & Technologies
04. Featured Work
Canvas SIS Prep Tool
Desktop App · Published to App Store · Cross-Platform
A cross-platform desktop utility I designed, built, and independently published to the Apple App Store. Built to eliminate manual errors in Canvas SIS CSV creation for educational institutions — adopted by the Workforce Training team at Salt Lake Community College to automate a previously error-prone course shell creation workflow.
Lorekeeper
Agent AI & LLM Orchestration
Full-stack JS application orchestrating Claude, Gemini, and ChatGPT through a Retrieval-Augmented Generation memory layer. Manages local AI environments (KoboldCpp, Ollama) to stress-test context windows and multi-backend system architecture.
The Plant Doctor
Machine Learning & Computer Vision
End-to-end image classification app using ML.NET and CNNs. Ran the full pipeline — data collection, ETL processing and sanitization via Jupyter, model training — through to a deployed classification interface.
05. What's Next?
Get In Touch
I'm actively exploring opportunities in AI engineering, BI architecture, and data-driven product development. Whether you have a project, a role, or just want to connect — I'd love to hear from you.