I was hired by Dr. Auguste El-Kareh (author of Electron Beams, Lenses, and Optics, 2 volumes), the president and founder of Alcedo, to develop an electron beam tester for finding defects in multichip modules (MCM), PCBs, or flat panel display substrates. This project was in partnership with Dr. Hans Pfeiffer's group at IBM Hopewell Junction.
I learned that the pain of timing is one of the most vital factors in the ability of a startup to survive its infancy. Dr. El-Kareh and the IBM team had great vision but unfortunately, were 10 years ahead of market needs. After 2 software managers were let go in a few months, I became the system manager to cover the software portion as well. I became a legend in the company after I managed to dump the 3 inch thick legacy C codes (user interface and electronic control from the consult computer) and replace them with a dozen basic building blocks in order for the user interface to communicate with the embedded Intel 80188 microprocessor and several electronic boards; the software project then moved forwardElectron microscopy column, induced current electron detectors, high/ultra high vacuum.
I was the only person at that time who could integrate a complex system which included 2 in-house control cards (with Intel 80188 micro-processors) for image formation/vector scan, control electronics for the electron microscopy column, and vacuum. I debugged Intel 80188 assembly with a logic analyzer.
By reading Intel 80188 assembly codes, PAL (programmable array logic) codes and numerous electronic diagrams, I managed to reverse engineer 2 in-house cards for image formation and electron beam vector scan.Wrote basic functions to communicate with the control electronics (2 in-house image formation/vector scan cards and one of-the-shelf ADC, DAC, digital card).
Modified the assembly code, which ran on the Intel 80188 micro-processor in the pattern generator card.The company had less than ten people and a high turnover rate. Because so much knowledge was lost, I had to reverse engineer 2 in-house electronic boards and assembly codes. In order to get the project moving, I wore other hats as well, including those of application, systems and software. This was my first startup experience. |