Engineering That
Ships.
Systems architecture, agentic AI, and infrastructure automation. 20 years of Fortune 500 engineering, now building in the open from the Colorado Rockies.
What I Build.
Production systems that survive contact with real data, real users, and real failure modes.
Infrastructure & DevOps
Sovereign infrastructure on bare metal. Self-hosted stacks, CI/CD pipelines, and monitoring — enterprise patterns without the enterprise tax.
- Bare metal on Hetzner + Coolify
- Self-hosted identity (Authentik)
- Automated deployment pipelines
- SRE monitoring & IOPS tracking
Agentic AI & Automation
Production AI systems that do real work — verification engines, data pipelines, and multi-signal orchestration at scale.
- LLM orchestration (local & cloud)
- Multi-signal verification engines
- Stealth browser automation
- Incremental persistence patterns
Systems Architecture
Enterprise-grade design at startup speed. State machines, fallback chains, and resilience patterns that survive production.
- Database-driven state machines
- Tiered processing pipelines
- Timeout stratification
- Failure-tolerant orchestration
Enterprise Thinking. Independent Execution.
"I build the systems I write about, and I write about the systems I build. The code is the resume."
20 Years of Enterprise
Mastercard, Jabil, Fortune 500 systems. The patterns that work at global scale, applied without the overhead.
Sovereign Infrastructure
Self-hosted on bare metal. Full control over the stack, the data, and the deployment pipeline. No vendor lock-in.
Built in the Open
The code is public. The blog posts show the real architecture, the real failures, and the real decisions.
Have an engineering challenge?
Whether it's architecture, automation, or agentic AI — I enjoy a good conversation about building things that last.
Get in TouchLatest from the Workshop
Engineering notes from the field — architecture decisions, production failures, and the real work behind agentic AI systems.
The Verification Layer Is the Product
I ran my production KYB pipeline at FP16, INT8, and INT4. Accuracy held at 92% across all three. A LoRA fine-tune dropped it to 12%. The model wasn't the variable.
The Three Enterprise Layers Are Collapsing Into One
For twenty years, enterprise software separated CRM, orchestration, and back-office into three layers with three vendor contracts. A confidence-gated decisioning layer is collapsing all three — and the vendors know it.
Quantization, LoRA, and the 8% Problem: Benchmarking Local LLMs for Production AI
I benchmarked 7 model variants across 3 quantization levels, tested a 3B challenger, built a self-training pipeline, and attempted QLoRA fine-tuning. The fine-tuning failed catastrophically. Here's what I learned.