An applied lab. I test how AI changes the way companies work, on real teams and on my own system. I also test open models on European clouds. Built fast. Kept only if they work.
Things that started here and now have real people using them. Each one lives on its own subdomain until it earns a name of its own.
An assistant for dental and orthodontic practices.
live pilotOne MCP server that grew into the connector for every Claude account in a small company.
in useWhere Berlin could still build homes: empty sites, low rooftops and single-storey supermarkets.
liveBerlin’s €45 billion budget in 4,500 dots: where the money comes from, where it goes, every line searchable.
liveAsk Claude who you talk to most, who you are drifting from, and who you are about to meet. Reads the graph your Mac keeps for Siri. Runs on your Mac, read-only.
open sourceSomething is always in the works.
brewing
The live site, captured 23 October 2026. The cuts shown are pia’s own example. How pia was built.
Open pia.supercharged.shIt started as a small MCP server so Claude could read one large, old FileMaker database. While building it I kept adding: writing that asks first, more of the company’s services behind the same connector, and the permissions from one central service, which now decide who can read, write or run what. Today it is the connector for every Claude account in a small company.
$ tools/list
list_tables read
query read
create_record write, asks first
update_record write, asks first
run_script run, asks first
add_fields schema, full access
The first tools, from the database it started with. Each call still runs on the signed-in person’s own login.
The lab notebook. Every experiment states its pass rule before the run, changes one thing at a time and uses invented data or my own. A result shows up here first, then as a finding with its data.
01 · Personal
E-02
E-01
02 · Shared
E-04
E-05
03 · Company
E-03
Teams I have worked with and learned from
Models change every few months. What a person, a team or a company knows, and can put into words, stays. The lab studies how that knowledge is found, said, agreed and kept current.
An AI can only use what it can see. Most of what someone knows is said out loud long before anyone writes it down. So I start with the people.
01 · Find
Map it where it lives. Nothing moves.
02 · Say
Most knowledge is said before it is written.
03 · Agree
The AI gathers, sorts and proposes. People decide.
04 · Keep current
When something changes, the old version goes.
Supercharged in 30 walks one person through it in 30 days. Supercharged Agents does it for teams and the company.
They follow the layers of the Supercharged Approach: one person, a team, the whole company.
01 · Personal
02 · Shared
03 · Company
Health and patient data should stay in Europe. So I test open models that run on European clouds.
If your knowledge lives in your own files, the model is a setting you can change.
Early observations, each from one team or one system. They shaped how I work. They are not results yet: a result needs a run with enough data behind it, and goes up in Running now first. Client names withheld.
F-01 · Personal
F-02 · Shared
F-03 · Shared
F-04 · Company
F-05 · Company
F-06 · Shared · Preliminary
01
Numbered notes from real teams and from my own system, on this page.
02
03
04
Setup, cases, the result in numbers, and the files to reuse. Otherwise it stays a field note.
Running beats perfect. Every experiment gets real users early.
Narrow tools that do one job well, not one assistant for everything.
If it doesn’t work, it goes. If it does, it graduates.