Skip to main content
Writing

Writing

Case studies and blog posts on architecture, operations, and delivery: what worked, what failed, and what is reusable.

Product OverviewOpen PlaygroundContact

Case Studies

Real implementations with metrics, stack details, and lessons learned.

All case studies →
Building a Multi-Model LLM Platform on Consumer GPUs
AI InfrastructureFeatured

Building a Multi-Model LLM Platform on Consumer GPUs

How I run multiple OpenAI-compatible LLM endpoints on a small K3s cluster with AMD Radeon GPUs, and what I had to do to make it stable.

24GB
GPU VRAM
4
Models Served
OpenAI API
Interface
KubernetesvLLMAMD Radeon 7900 XTXPrometheus+2 more
Read case study
fi-fhir: Format-Agnostic Healthcare Integration
HealthcareFeatured

fi-fhir: Format-Agnostic Healthcare Integration

A case study in healthcare integration: how Source Profiles, a three-phase parsing pipeline, and a workflow DSL turn messy legacy formats into semantic events.

5+
Formats Supported
70-90%+
Parser Coverage
Config-first
Integration Time
GoYAMLCELSvelteKit+8 more
Read case study
Healthcare API Integration: Patient Matching & Documentation Gaps
Healthcare

Healthcare API Integration: Patient Matching & Documentation Gaps

A real integration failure mode: when an API contract is implicit, patient matching becomes a risk management problem. Here is the defensive pattern (validation + contract tests + instrumentation) that makes it operable.

99%+
Match Accuracy
~2M/mo
API Calls
-75%
Debug Time
FHIR R4HL7v2GoPostgreSQL+2 more
Read case study
Building a Real-Time Kubernetes Operations Dashboard
DevOps

Building a Real-Time Kubernetes Operations Dashboard

How I built FlexDeck, a full-stack operations dashboard with real-time K8s monitoring, GitLab CI/CD visualization, and AI model management using Go and SolidJS.

<100ms
Update Latency
3
Clusters
99.9%
Uptime
GoSolidJSKubernetesD3.js+2 more
Read case study

Blog

Technical notes, implementation updates, and deeper dives.

All posts →
Getting Gemma 4 Running on a Radeon 7900 XTX (with and without TurboQuant)
April 4, 2026·8 min read
Lab

Getting Gemma 4 Running on a Radeon 7900 XTX (with and without TurboQuant)

What it took to get Gemma 4 E4B serving cleanly on Radeon through FlexInfer: a stable TRITON lane on a 7900 XTX, an experimental TurboQuant long-context lane on a second node, and the GPTQ pipeline work still underway.

gemma4amdradeon7900xtx+6 more
Read post
Build Your Own Legs Before the Crutches Fail
March 9, 2026·8 min read
Professional

Build Your Own Legs Before the Crutches Fail

AI-assisted development is useful leverage, but only if you convert borrowed competence into real judgment before the support becomes a dependency.

ai-assisted-devengineeringagentsdeveloper-workflows
Read post