DeveloperOrchestratorSocratic thinkerQuarterback
I run the agentic loops behind physical AI. As Quarterback of an Amazon Robotics fulfillment floor, I keep hundreds of Drive Units moving, amnesty monitors clean, and the whole orchestration humming - then I build the software that makes that floor legible in real time.
I sit where physical machines meet software intent.
My day job is Quarterback at an Amazon Robotics fulfillment center - the person accountable for floor health when the robots are running live. That means reading Drive Unit behavior, clearing amnesty monitors before they cascade, and steering recovery so throughput never stalls. It is hands-on systems thinking under real-time pressure.
Off the floor I write the tooling that makes all of it legible. I am drawn to physical AI and multi-agent systems - the orchestration problem of getting many autonomous actors, human and robotic, to converge on a clean state. I treat every incident as a root-cause question, not a symptom to mute, and I build software with the same discipline: small, observable, and honest about what it does.
A sleek desktop dashboard that pulls the whole floor into one pane: live warehouse metrics, NSTA tracking, and robotic navigation laid out so a Quarterback can act in seconds. Built native-feeling and fast, it turns scattered telemetry into a single source of truth for the people running the floor.
Live metrics
NSTA
Robotic nav
“Don’t wish it were easier. Wish you were better.”
That line sits behind how I work. The Socratic move is to keep asking why until you reach the real constraint - then raise your own standard to meet it. On the floor and in the codebase, integrity means owning the root cause instead of dressing up the symptom.
Building in physical AI, robotics ops, or agentic systems? Let’s compare notes.