After four years of EE, I joined GE Space Systems, Valley Forge (now Lockheed Martin). I did a Master's part time at U. Penn as part of the GE Advanced Course (very tough, but highly recommended if you are up to it). Among other things, I worked on microwave measurement systems and microwave circuit analysis software.
After four years there, I moved to GE Syracuse Electronics Laboratory Syracuse (later sold to BAE and moved to Nashua, NH). There I worked some of the first GaAs monolithic microwave integrated circuits (MIMICs). Cool stuff, but the tuning and tweaking common with microwave circuits was no longer possible. Doing six or more design iterations was way too length and costly. So I went to Syracuse University full time for my Ph.D. in electromagnetics. My advisor was Roger Harrington, father of the Method of Moments, which is in wide use in microwave electromagnetics, but at that time, mostly for antennas.
There, I came up with a neat way to apply MoM to multilayer planar circuits. When I completed my Ph.D. I tried to find a company that would commercialize the technique. No luck. I also tried at both SU and Cornell for a tenure track position. Again, no go. Both outcomes really bummed me out, but are probably among the best things that ever happened to me...because I decided to take the plunge and start a company and commercialize it myself.
That was 30 years ago. We now have 20+ employees and we are viewed as the standard for accuracy in a now crowded field of applied numerical microwave electromagnetics. Wow, what a ride that has been!