THE FOUNDATION
Infrastructure
Servers, networks, and systems administration. I learned to look beneath the interface and understand what keeps a system alive.
VOLUME 01
2007—NOW
更新中 / IN PROGRESS
2007—NOW
更新中 / IN PROGRESS
A field notebook by Jesse Anderson
The road
is the work.
I’m a software engineer who took the long way here—through server rooms, cloud architecture, product code, and now AI. This is where I keep the things I’m building and what they’re teaching me.
Still learning. Still making.
Usually the hard way.
CHAPTER ONE / THE LONG WAY AROUND
How I got here.
THE TURN
Software
A self-taught path into professional development. Building became the way I learned—and the way I made ideas useful.
THE WIDE VIEW
Architecture
A year as a Solutions Architect taught me to see the whole landscape: people, constraints, tradeoffs, and the connections between them.
THE OPEN ROAD
AI & tools
At Disney, and in my own work, I’m exploring how people can build with capable agents without giving up understanding or responsibility.
遠回りにも意味がある
There is meaning in the long way around.
CHAPTER TWO / CURRENT EXPEDITIONS
What I’m carrying forward.
LOCAL AI AGENT COMMAND CENTER
Cyberdeck
“Code became cheap.
Understanding did not.”
A keyboard-first home for running multiple local coding agents without losing track of the work—or my responsibility for it.
Why I’m building it To understand what dependable agent systems actually need: identity, visible state, honest permissions, and an operator who remains in control.
A CLIMBER’S TRIP NOTEBOOK
ClimbSite
A trip planner connecting climbing areas with camping, approaches, parking, and the practical details that shape a weekend outside.
Why I’m building it Because the best days outside begin long before the first climb—and useful local knowledge is usually scattered everywhere.
Open the project ↗A SMALL LAB FOR AGENT MECHANICS
Agent
Workbench
Take it apart.
Put it back together.
Understand every piece.
A deliberately small Python lab for learning agent loops from the inside: tools, structured outputs, evaluation, and safe boundaries.
Why I’m building it I don’t want to understand AI only from the surface. This is the notebook where I test the machinery.
Open the project ↗AN AUDIO STUDIO FOR JAPANESE LEARNING
Shoshizutsu
Studio
少しずつ
A little at a time.
A local Python studio for making relaxing Japanese-learning podcasts with distinct AI-generated voices, ambient sound, and timestamped Japanese and English transcripts.
Why I’m building it To explore how local AI can support language practice while keeping script review and audio quality part of the process. Generated scripts need explicit approval before they become episodes.
Ask about the studio ↗CHAPTER THREE / NOTES FROM THE ROAD
Learning, recorded.
Not thought leadership. Just an honest record of decisions, dead ends, things that worked, and what I understand now that I didn’t before.
01
FIELD NOTE
↗02Documenting the work again
Why I’m making space to record the experiments, decisions, and lessons behind what I build.
FIELD NOTE
↗Why I built Cyberdeck
Code became cheap. Understanding did not. Cyberdeck grew from my need to manage multiple coding agents without losing track of the work—or my responsibility for it.
A NOTE FROM JESSE
I’m not trying to arrive.
I’ve spent nearly two decades following my curiosity through technology. The labels have changed—systems administrator, developer, solutions architect, software engineer—but the habit underneath has stayed the same: get close to the machinery, ask how the pieces connect, and make something that helps.
This site is a record of that journey while it is still happening. If something here connects with what you’re working through, I’d be glad to hear from you.