Turning industrial
complexity into
streamlined intelligence.
Four years of SMT process, AOI, and flying probe test on a PCBA line. I build the tooling that turns what those machines emit into decisions, and I write about it weekly.
Recent writing
The Model Finds the Defect. The Standard Decides What It Means.
A local YOLOv8 + offline LLM pipeline that maps PCB defects to the right IPC standard and process cause, plus the regex firewall that keeps bare-board and assembly domains from blending, and the recall floor it never beat.
The Co-Pilot That Refuses to Guess: Grounded RAG for PCBA Test Logs
A synthetic PCBA test-log pipeline, SPC that runs itself, and a citation-forced RAG co-pilot that refuses to answer when it isn't sure — built over eight weeks of evenings and weekends.
The Tool Was Never the Point: A 12-Step Workflow for AI-Native Projects
I built an AI co-pilot for PCBA test analytics to fill a portfolio gap. The tool became the least interesting thing I made. The real deliverable was the 12-step workflow I used to govern the agents that built it.
Stop Applying Tape: RCCA in Agentic Debugging
AI agents are good at finding something that works. They are terrible at finding out why something broke. That is a discipline problem — and manufacturing solved it sixty years ago.
The Hidden Tax on Your AI: Context Windows, Skills, and the Cost of Attention
Every skill you load, every message you send, every document you attach quietly eats into the same fixed budget. Here's what happens when that budget runs out — and how to spend it wisely.
Running Multiple AI Agents in Parallel Without Destroying Your Codebase
Two AI agents writing to the same file simultaneously is a data corruption event. Here is the file-locking orchestration layer that makes parallel AI development safe and fast.
// arcade · built for this portfolio
I built a game.
You are a signal packet navigating copper traces on a live PCB board. Collect every solder joint to clear the board — but four rogue AI agents are hunting you, each level smarter than the last. It's Pac-Man if Pac-Man cared about manufacturing.
// enemy roster
Through a cinematic lens
// about
Manufacturing Engineer.
AI-Native Builder.
Photographer.
I run SMT process, AOI, and flying probe test on a PCBA line in Dallas — MyData pick-and-place, Mirtec AOI, Seica flying probe, Heller reflow, IPC-A-610 Class 2. Four years of it. Then I build the Python and SQL tooling that turns what those machines emit into something an engineer can act on.
That combination is the point. A Lean audit I led lifted throughput 20%. Standardizing PCB electrical test protocols cut rework 10%. Assembly Hub, an internal tool I built by directing AI agents rather than writing the code myself, has run 120+ PCB assembly jobs in production and is used daily by four operator roles.
I write about where those two sides meet: grounded RAG for test logs, YOLOv8 defect classification mapped to IPC standards, agentic workflow design. Off the floor I'm usually behind a camera shooting stories through a cinematic lens. More about me or my projects.