From 1988 to 1998 I worked for Fokker Space in Amsterdam and Leiden. These were formative years for me.
In a way, after a few exhilarating years in the flight test department of Fokker Aircraft, this almost felt like my first "proper", professional job.
The people there were true intellectuals, like I have never experienced since. Most of them held PhD's in astronomy, and had broad interests. One thing they were not though, was engineers. The whole thing was an object lesson in how far you can get with basic knowledge, without having learned engineering design. It turns out you can come a long way, but at a price.
Refer mostly to configuration control now.
My first assignment at Space was to write and run the software simulations for the sensor and actuator hardware, and a prototype of the on-board software, of ESA's astronomical satellite ISO, and I spent the best part of five years doing that. Out of this grew my work on a new, non-saturating "proximal" control law, which became the so-called "dual control law" which later flew on the satellite.
Another spin-off for me was the extensive use of the quaternion in the equations for the satellite's motions. Getting an intimate feel for the care and feeding of quaternions has served me well ever since.
I then moved over to consultancy work for the SAX satellite and a few smaller control projects. The SAX satellite gave rise to some work on the the somewhat arcane Kalman filter, which I mastered enough to write a useful manual on how to use this in the satellite's no-gyro control mode.
After the satellites, I spent the next five years in the zero-G department, which was a fun playground for someone more interested in flying than in spacecraft. See the separate section on the zero-G department.