I Build Human-Centered AI Products
AI Product Management & Applied AI Research | Multi-Agent Systems | MS Candidate, Human Centered Computing
AI-powered Products that optimize for model capability over human intent tend to fail. I'm focused on closing that gap through product, research, and educating others.
- 350MCP commands behind one command surface in SpawnForge, so a person and an agent drive the editor the same way
- 30+MCP integrations feeding Ember's agents real-time context, in private alpha
- 3model families in one scaffolding decision loop, built for the MS thesis and awaiting IRB approval
- 20M+events a day through the platforms I led across 12+ years at Werner Enterprises
Scope I led at Werner Enterprises, 2021 to 2026
- sevenproduct and UX leads, each owning strategy for their own teams
- 13agile teams that product direction covered
- 150engineers in those teams, roughly
Working together
I take on a small number of advisory, consulting, speaking and coaching engagements alongside my full-time work. Here is what each one looks like.
About Me
I'm a Sr. AI Platform Product Manager and an MS candidate in Human Centered Computing at the University of Nebraska Omaha. My thesis measures AI-driven adaptive scaffolding against a fixed faded schedule of worked examples, and what each does to cognitive load and skill acquisition; the protocol, the instruments, and the multi-agent scaffolding system are built and awaiting IRB approval. I build systems to solve problems and gaps that I see in the world: Ember, SpawnForge, Frets, Player2 Kanban... Ember runs on an OWA (Orchestrator-Worker-Antagonist) architecture: a worker agent produces, an antagonist agent adversarially evaluates the output for authenticity and accuracy and sends it back until it passes, and an orchestrator makes the final call, with each role on the model that suits it. SpawnForge is an AI-native game engine in the browser whose 350 MCP commands sit behind a single command surface, so a person and an AI agent drive the editor through the same code path. I've helped onboard 300+ developers, product managers, and IT staff to AI coding tools, worked with technology leads across 20+ product engineering teams company-wide to shape AI-enablement strategy, and guest lectured at UNO on IoT, cloud infrastructure, and AI adoption.
That work rests on 12+ years at Werner Enterprises, where I progressed through 11 different roles from operations to Sr. Product Manager. In that last role I owned product direction for 13 agile teams and roughly 150 engineers, through a product and UX team of seven who each owned strategy for their own teams. Drive Werner Pro serves 10,000+ drivers daily for scheduling and compensation management. I led the migration of legacy electronic logging devices (ELDs) across 8,000+ tractors to a cloud-integrated Platform Science system, replacing aging infrastructure. I developed a maintenance appointment system that generates $5M+ annual savings by optimizing fleet routing and asset utilization, and Maintenance Connect, a mobile inspection application, eliminated 60,000 sheets of paper annually across technician workflows. My team built the first fleet with direct integrations to all major Class 8 OEMs in the US, and I led technology partnerships with frontier AI and autonomous vehicle companies (Google, Kodiak, Aurora) to shape their product roadmaps and secure exclusive early access to emerging integrations.
I've lived the problems I solve, such as isolated systems, missing context, and speed-bumps to throughput, and I develop leaders: my team members have gone on to take roles like AI Product Manager and Executive Director of Technology, winning awards and earning their own credibility along the way.
Expertise
Certifications
Recognition
- Samsara Technology Leader of the Year, Finalist (2023)Technology leadership in fleet operations, in the Connected Operations Awards.
- Supply & Demand Chain Executive Top Supply Chain Projects (2022)The Event Resolution System, a product my team built.
- UNL Senior Design Showcase, Platinum Award (2022)The Trailer Tracking and Weather app.
- HDT Emerging Leaders (2020)The Werner Edge rollout: driver dwell time cut in half and training time down 60%+. One of five leaders recognized nationally.
- Werner Recognition Awards, four consecutive years
Key Achievements
AI Systems & Research
- Four systems built: Frets and Player2 Kanban are publicly usable, Ember is in private alpha, SpawnForge is in active development
- 350 MCP commands behind a single command surface in SpawnForge, so a person and an AI agent drive the editor through the same code path
- 73 visual scripting node types in SpawnForge that compile to TypeScript
- 30+ MCP integrations give Ember's agents real-time context from research databases, financial APIs, and web search
- An OWA architecture (Orchestrator, Worker, Antagonist) running three model families in one decision loop, which I designed and built
- MS thesis instrument built and awaiting IRB approval: protocol, measures, and the adaptive scaffolding system
- Player2 Kanban published on npm, with an MCP server that AI coding agents connect to directly
- Frets runs live on this site on an open TypeScript theory engine, fully client-side
- Helped to onboard 300+ developers, product managers, and IT staff to AI-powered productivity & coding tools
- Shaped AI-enablement strategy with technology leads across 20+ product engineering teams
Enterprise Delivery Record at Scale (Werner Enterprises)
- Directed products processing 20M+ daily events in a 24/7/365 operation
- Shipped Drive Werner Pro to 10,000+ drivers for scheduling, compensation, and career management
- Shipped Maintenance Connect to technicians, eliminating 60,000 sheets of paper annually
- Shipped Breakdown Management System (evolved into Event Resolution System) handling 250,000+ maintenance events annually
- Developed a maintenance appointment system that optimizes fleet routing and asset utilization
- Led ELD migration across 8,000+ tractors from legacy systems to cloud-integrated Platform Science
- First fleet with direct, bidirectional integrations to all major Class 8 OEM maintenance systems in the US
- Direct partnerships with frontier AI and autonomous vehicle companies (Google, Kodiak, Aurora) to shape product roadmaps
- Reduced driver dwell time by 60%+ through workflow optimization
- Reduced employee training time from 3 weeks to 3 days through improved design and processes
- Financial impact: $5M+ a year saved through appointment-based maintenance, $1M+ a year in warranty filings that previously went unclaimed, millions more a year from retiring AS400 onto cloud, and fleet maintenance turned from a cost center into a revenue generator through managed services for external carriers
- Led a product and UX team of seven, each owning strategy for their own teams, covering 13 agile teams and roughly 150 engineers
- Team members have gone on to roles like AI Product Manager and Executive Director of Technology
- Developed Product Management interns alongside the permanent product and UX leads
Research: Human-Centered AI Systems
My research explores how to synthesize human cognitive needs with model capability, improving the outcomes achieved when we collaborate with Artificial Intelligence. I study the intersection of cognitive load theory and adaptive learning interfaces, focusing on how we can build AI systems that enhance human capability without creating dependence.
Current Thesis
The Effects of AI-Driven Adaptive Scaffolding vs. Static Worked Examples on Cognitive Load and Skill Acquisition
Research Interests
Currently Exploring
- Multi-agent architectures for quality control and authenticity (Ember)
- Cognitive load in human-AI systems
- Adversarial model evaluation to prevent bias and hallucination (how I evaluate model output)
- Adaptive learning interfaces for AI assistance
- Autonomous vehicle integration strategies
- AI at scale in regulated industries
I write about multi-agent architecture and what changes for product teams when models do more of the building. Substack is where I publish now, and the earlier articles are on LinkedIn.
Dogs
A lesson learned from a hairstyling instructor, applied to the agentic era.
June 2026
Picks and Shovels
In a world where anyone can build anything, what will you build?
May 2026
Introducing: The OWA Agent Architecture
A scalable framework for reducing AI slop in practice and production.
March 2026
Earlier articles on LinkedIn
View all- Teaching a Machine to Think: An Introduction to Explainable Active LearningResearch · January 2026
- The Cost Curve of Experimentation: How AI Enables Ideation and MVP Development at ScaleStrategy · January 2026
- Building AI Agents With AI: A Framework for Multi-Agent Software DevelopmentFramework · October 2025
- Building Your First AI-Agent Team: A Beginner's GuideTutorial · October 2025
- 30 Minutes a DayGrowth · September 2025
- The Rules Haven't Changed, The Tools HaveStrategy · January 2025
Featured Projects
View AllSystems I built and products I led. The AI work is open and inspectable: public repos, and Frets runs live on this site. The enterprise work shipped into a 24/7/365 operation processing 20M+ daily events.
Ember
Ember is a multi-agent system in private alpha, wired to 30+ external services over MCP (Model Context Protocol), giving its agents real-time context.
SpawnForge
SpawnForge is an open-source, AI-native game engine that runs in the browser, with 350 MCP commands that let you or any AI agent drive the editor.
Autonomous Truck Partnerships
Led the partnerships that integrated autonomous trucks into Werner's operations: TuSimple, Embark, Kodiak, and Aurora publicly, plus others under NDA.
Press & Recognition
View AllFeatured in industry publications and recognized for AI & transportation leadership.
Awards & Recognition
Featured In
Creative
Music
More than Blood
Bass
I played bass in this band. We put out a few recordings that are still on Spotify - a fun reminder of my creative side outside of product management.
Nevermind the Riot
Lead Vocals
I was the lead vocalist for this project. These recordings go back a bit further but still capture some memorable moments from that era.
Lesser Evil
Vocals & Guitar
My first band. These are live recordings from way back when - raw energy and the excitement of performing for the first time.
Tony Gebhard
Featured Vocals
A collaboration with Tony, a talented solo artist who traveled from Ohio to Omaha for the More than Blood CD release party on his way to Alaska.
BBQ
BBQ_Reveal
YouTube Channel
I cook on three Big Green Eggs, making everything from brisket to octopus. I enjoy trying new recipes and tweaking old favorites, then sharing great food with family, friends, and neighbors. This channel is fairly casual - a good friend and former colleague helps me manage it.
Watch on YouTubeLet's Connect & Exchange Ideas
Topics I'm always exploring:
- •Multi-agent AI architectures for quality, authenticity, and real-world application
- •Human-centered AI systems (research + product + implementation)
- •Fleet technology & IoT at scale
- •Teaching & mentoring the next generation of product leaders
- •Adjacent domains (creative work, emerging tech, cross-functional problem-solving)
I'm constantly learning and engaging others to share knowledge. I am interested in intelligent conversations around emerging technology, product strategy, autonomous systems, and agentic workflows that respect human intent. Whether you are working on frontier AI, self-driving technology, robotics, or just want to share ideas, I'm here for it.
Get in Touch