Why I'm Building in Public: From Home Lab to Production AI

I've been building things with technology for over 23 years. For the last decade-plus at Microsoft, I've had the privilege of working as Chief Data Officer (CDO) for Federal Civilian LLS ATU, helping organizations transform how they use data and AI.

But here's the thing — some of my best learning happens at home, in my lab, where I can break things without consequences.

The Home Lab

What started as a single server has grown into a 3-node Proxmox cluster with Ceph distributed storage, 95+ Docker containers, and a full enterprise-grade monitoring and security stack. It's where I test ideas before recommending them to customers, and where I push technology further than any corporate environment would allow.

Here's the high-level architecture:

• 3 Proxmox nodes with Ceph providing replicated block and file storage across all nodes
• 95+ Docker containers running everything from media servers to AI workloads
• Full observability with Grafana (34 dashboards), Prometheus, Loki, and Alertmanager (40+ alert rules)
• Enterprise security including Wazuh SIEM, Suricata IDS, vulnerability scanning, and CrowdSec
• Local AI inference with 34+ LLM models on consumer GPUs

Why Build in Public?

There's a gap in technical content. Most blog posts are either too basic ("how to install Docker") or too theoretical ("the future of AI"). I want to share the messy middle — the actual implementation details, the debugging sessions, the architecture decisions that only make sense after you've tried three wrong approaches first.

Every post on this blog comes from real infrastructure I operate. When I write about monitoring, I'm sharing patterns from dashboards I check daily. When I write about security, it's from a stack actively defending real services.

What to Expect

This blog covers the intersection of enterprise technology and hands-on engineering:

• Azure & Cloud Architecture: Real deployment patterns, cost optimization, and lessons from production systems
• AI/ML Engineering: Local LLM inference, RAG pipelines, agent frameworks, and practical AI applications
• Data Engineering: Pipeline design, database optimization, and data platform architecture
• Infrastructure & DevOps: Container orchestration, monitoring, security hardening, and automation
• Home Lab Projects: The proving ground where all of this comes together

The AI Pipeline Behind This Blog

In the spirit of building in public, I should mention that this blog itself is partially powered by an AI content pipeline I built. An n8n workflow generates draft posts from topics I queue, publishes them to Ghost CMS for my review, and on approval syndicates to LinkedIn. The entire pipeline runs on my home infrastructure — local LLMs for ideation, cloud APIs for polish, and automation for distribution.

It's infrastructure all the way down.

Let's Connect

If you're interested in Azure, AI, data engineering, or just want to talk shop about home lab builds, I'd love to hear from you. The best ideas come from the community, and I'm always learning something new.

Welcome to Build Without Bounds. Let's build something.