Why This Journal Exists
I have spent more than 20 years as a software engineer. I do not want the next stage of my career to be a reset. I want it to be an extension.
The goal is to bring software architecture, distributed systems, cloud infrastructure, reliability, deployment, observability, data systems, and debugging into robotics and Physical AI.
For roughly the next two years, the objective is adaptation: become capable of independently building, integrating, debugging, and operating real robotics systems rather than only consuming tutorials or vendor demos.
The longer horizon is about ten years. I want this repository to survive the beginner phase and continue through professional Physical AI engineering, possible consulting work, larger systems, and eventually the lessons that come from having done this for years.
That is why the early entries matter.
I do not want the history to begin only when the work looks impressive.
The record should include questions such as:
- What is this connector?
- Why is the battery interrupting development?
- Why is VNC the wrong abstraction?
- What does the vendor ROS stack actually do?
- What should run on Jetson versus an RTX 4090?
Later, those questions should become things like:
- How should this autonomous system recover?
- How should a robot fleet be observed and updated?
- Where should inference execute?
- How should multimodal agents interact safely with robot capabilities?
- What architecture works in a real Physical AI system, and what failed?
What I want the evidence to show
The useful evidence will not be certificates or a list of tutorials completed.
I want to accumulate systems I can point to and say:
- I built this.
- I broke this.
- I diagnosed this.
- I measured this.
- I redesigned this.
- I understand why it works.
The public repository is the curated record. Raw command output, sensitive configuration, vendor-specific investigation, and unfinished work stay in a separate private research repository until they are worth publishing.
The goal is to make documentation part of the engineering process rather than a second job.
This is the starting point of the record.