I Do Run AI Agents Overnight. Here's What Actually Matters.
Boundaries, supervision, evidence, and why unattended execution is not the same thing as unbounded autonomy.
Petie Clark — Applied AI Systems
Applied AI systems that survive contact with the business.
Field notes on agent orchestration, evaluation, authority boundaries, enterprise integration, deployment, and production operations.
Get new posts in your inbox
No spam. No marketing. Just the writing.
/start-here
How I think about governed systems
Why I Prefer One Clear Source of Truth
Why tools can multiply while authority, state, and accountability still need one dependable home.
Why I Write Down the Rollback Before the Change
A deployment is not ready merely because the forward path looks convincing.
The Case for a Smaller Blast Radius
Production systems become trustworthy through bounded change, not theatrical confidence.
Why HR System Integrations Are Still a Dumpster Fire in 2026
After a decade building integrations with ADP, Workday, Okta, and Entra ID, I've learned that SCIM is a guideline, not a spec. Here's what actually works.
3 min read
Read →
My Homelab Grew Up: From Hobby to Production Infrastructure
Three years ago my homelab was a Raspberry Pi running Pi-hole and a Plex server. Today it's a multi-node cluster handling production workloads for paying clients.
2 min read
Read →
Fine-Tuning a 20B Model on 512GB Apple Silicon: What I Learned
I fine-tuned a 20-billion parameter model on my Mac Studio M3 Ultra and took validation loss from 1.85 to 0.79. Here's what I learned about running serious AI workloads on Apple Silicon.
3 min read
Read →
The Integration Engineer's Dilemma: Build vs. Buy in 2026
I spend most of my professional life connecting systems that were never designed to talk to each other. After doing this for years, I've landed on a framework
2 min read
Read →
What I Learned Building a SaaS Platform in a Weekend
Last month I built a full-featured SaaS platform in about 48 hours. Not a landing page. A working application with auth, multi-tenant data isolation, Stripe billing, and a
2 min read
Read →
Why I Run AI Models on My Own Hardware
I've been running large language models locally for a few months now, and honestly? I'm not going back to API-only workflows. The Cost Math Cloud
2 min read
Read →
From Side Projects to Scaling Up: Lessons I’m Carrying Forward
🧠 Context / Why I Did This For the past few months, I’ve been building Steadfast , a personal productivity app that’s been equal parts fun, frustrating, and eye-opening. Like
2 min read
Read →
App Prototypes from the Terminal: 5 Christian Tools I’m Exploring for Real-World Impact
🧠 Context / Why I Did This I've always believed that tools should serve people, not the other way around. And as a developer who's both deeply technical
3 min read
Read →
Baseline Setup, Part 6 – Automating the Rebuild with Git and Ansible
The final piece of the rebuild: how I use Ansible and Git to automate service deployment, node setup, and config management across my homelab—so I can rebuild fast when it matters.
2 min read
Read →
Baseline Setup, Part 5 – Service Visibility with Homepage
See how I deployed Homepage in Kubernetes to create a clean, fast, and live dashboard for my homelab. From grouped services to container health, it’s my daily launchpad.
2 min read
Read →