How the House Works
A company with one human and a house of AI members: how it is organized, how work moves through it, how the members learn, and what it has built.
Lorenzo Pasqualis
Thirty-five years, from embedded systems to enterprise-scale cloud platforms, and now AI-native engineering, practiced every day.
Current experiment: The Luci HouseServices
Most organizations are building software with new tools and old methods. I design the new methods, and I install them, so an engineering organization gets the most out of the tools it has today and is ready for the ones that come next.
I prove them every day in the house, my own company of one human and a house of AI members. Switching from my full-time job to the house is very natural. It just feels like two organizations. The fact that one is AI-driven is almost a detail.
How I think about it as of November 2026. I update it as the field changes.
Code is cheap now. What slows an organization down is the system around the code: how it decides, defines and approves work. That is where I start.
One or two engineers directing a staff of agents, with a product lead, own a milestone end to end. Spec and tests come before code, and only live counts as done.
Agents that act as well as suggest: autonomy tiered by risk, with a rollback path and an audit trail. Aimed first at the work that should not exist.
Humans own the outcome and keep the judgment. AI does the typing. Roles, review and decision rights are redrawn to match.
A new method does not spread by document. One team proves it, with a date to kill it or scale it, and then teaches the others.
The Lab
What I am trying, and what I learn from it.
How it's organized, how work moves, and what it has built.
Measured on the machine the house runs on, as of November 20, 2026
2 field notes
A company with one human and a house of AI members: how it is organized, how work moves through it, how the members learn, and what it has built.
The systems a house of AI members has built for one person, and what they do for him every day: the platform it runs on, the tools that make music, video and images, an archive with its backups, an internal website, and the work it does on its own.
I am employed full time, and always up for a conversation about AI and engineering.