Operational technology · Applied AI
Sr. Manager of OT Data Solutions at Williams. I own the governance and architecture that make pipeline and facility telemetry a trusted enterprise asset. On my own time I build applied AI systems for industrial data problems, hands on keyboard.
Figures 1 – 5
Applied AI systems I designed and wrote myself, aimed at problems I met in thirty years of operations. Each figure has its own sheet with the architecture and the decisions behind it.
Scanned engineering drawings to an auditable process graph
A nine-stage computer-vision and spatial-database pipeline that converts scanned piping & instrumentation diagrams into a queryable process graph: components, pipes, tags, and connectivity. The core extraction path uses no machine learning at all, and that choice drives the whole design.
A plain-English description in, a running verified application out
An autonomous software factory: a frontier model designs the app, a local LLM builds it chunk-by-chunk in a sandboxed test loop, failures feed a retrieval-and-fine-tuning learning loop, and nothing ships unless the assembled app actually runs.
Raw PLC tags to a Unified Namespace, with LLMs in the loop
A visual, node-based pipeline builder that puts language models to work on the hardest manual step in any Industry 4.0 program: deciding what every tag in every PLC actually represents. It ingests real Rockwell/Allen-Bradley exports, classifies equipment, generates descriptions, and proposes the namespace hierarchy.
A phone line that calls you about a critical alarm and takes your acknowledgment
A real-time voice assistant that answers phone calls, transcribes live, and responds conversationally. The part that matters: when a natural-gas facility raises a critical alarm, it calls a human, describes the condition out loud, and records the verbal acknowledgment.
A local-first LLM router with a judge-gated escalation path
A self-hosted, OpenAI-API-compatible router that answers every request with a cheap local model first, has a second model grade that answer against a rubric, and calls a paid frontier model only when the local answer measurably fails.
Additional work lives in private repositories. Happy to walk through architecture, code, and tradeoffs in conversation — just ask.
General notes
Most AI practitioners have never sat in a control room. Most OT leaders have never trained a model.
Industrial AI projects rarely fail on the modeling. They fail on assumptions about the data: that tags are clean, that timestamps align, that a historian value means what its name says, that the field device was actually reporting that day.
I work both sides of that gap. I've run the control-room systems that generate the data and the governance programs that make it trustworthy, and I build the systems that consume it. On one of the projects below, the right engineering call was to not use a model at all.
30+ yrs
Operational technology
Four SCADA teams across gathering, transmission, storage, and liquids. Multiple full system replacements sequenced to hold control-room availability through every cutover.
Hands on
Applied AI practice
Computer vision, locally hosted language and vision-language models, real-time voice pipelines, multi-provider inference routing, MCP tool interfaces. TensorFlow Developer certified. I write the code.
Enterprise
OT data governance
Led the MQTT edge-connector and unified-namespace rollout streaming high-fidelity field data to enterprise analytics in real time, inside OT security boundaries.
Service record
Control systems supporting 24/7 pipeline operations — four companies, every scale from field technician to multi-team leadership.
Tulsa, OK
2009 — Present
Oklahoma City, OK
2005 — 2009
Houston, TX
2002 — 2005
Oklahoma City, OK
1996 — 2002
Earlier service
Electronic Technician
Mason & Hanger Inc.
1992 — 1996
Installed and maintained electronic security systems protecting nuclear assets. Served on the OSHA compliance team.
Telecommunications Analyst
Phillips Petroleum Co.
1990 — 1992
Installed and maintained conventional and trunking radio systems supporting field operations.
Sergeant — Avionics & Autopilot Technician
U.S. Air Force
1986 — 1990
Maintained inertial navigation and autopilot systems on AWACS Boeing 707 airframes.
Legend
Off the clock
Board Member
Industry association for energy telecommunications and data professionals.
Advisory Board Member — AI Curriculum
Shaping how the next cohort of engineers is taught to build with AI.
Instructor
Teaching measurement and SCADA practice to industry practitioners.
M.S., Energy Management
Oklahoma City University · 2016
B.S., Management & Ethics — Magna Cum Laude
Mid-America Christian University · 2002
A.A.S., Electronics & Programming
Clarendon College · 1986
TensorFlow Developer Certificate
Zero To Mastery Academy · 2023
Lean Six Sigma Green Belt
University of Oklahoma
NCO Preparatory School
Community College of the Air Force
Correspondence
Open to conversations about AI and data leadership roles in energy, manufacturing, and industrial software — and to comparing notes with anyone working the same seam.
