Worked for Wolfram Research, Inc. Champagin, IL. My specific group is the Release Engineering group. I worked as a full-time non-permanent employee, rather than an intern. The reason for this is that I already had a semesters worth of experience with them working on MacOS installer code. Not being an intern allowed them to have a whole different set of expectations for me. Most interns were doing research-type work; 98% of my job was related to releasing the next version of Mathematica for MacOS. The other 2% is reseach work that probably will not be realized 1) until I am working for Wolfram full-time, and 2) until the Fall release of Mathematica is past. Or it could just as well never be realized.Released or very-near-release products and what I did for them:
This is some information that I can actually talk about because these are products that have been publicised.
Worked for Motorola, Arlington Heights. The group was the Stability, Stress and Performance test group for Personal Digital Celluar Provider Products. Basically, I enhanced a program that generated messages between an OMC (operations & maintenance center) and a MM (module manager). The "OMC load tool" used a monolithic C++ API to transmit messages between machines with parameters such as linear or random message distribution over the time period and the number of messages to be sent per hour. There were 3 types of messages that could be sent: BusyIdle, events and caller data logs. It utilized realtime processes on an IRIX host for acurate message firing. The IRIX host was a SGI Indy emulating a MM, and the OMC was a Tandem Non-StopUX machine. The API also allowed for multiple MM's... but real MM's are also Tandems, and it would be quite expensive to use 7 Tandems when 7 SGI's would due... so another task of mine was to upgrade IRIX on a few SGIs, administrate, and install the "SuperCell API" libraries and binaries. I familiarized myself with some of the shared memory, multi-processor, and administration commands that NonStopUX and IRIX provide in the process of doing this task. Much of my work was mainly in upgrading the original load tool from an older version of the API to the most recent release as well as moving the source from the a real MIPS machine to the SGIs. At the end of the summer, I started work on a "performance monitor collector" program that would effectively try and stress out the performance monitor process on the OMC. There was not enough time to finish the 'migration' project because the API had changed so dramtically, and there was no API documentation and the routine names were pretty cryptic. One of the extra benefits I gained from this experience was that my first two weeks were spent reading detailed tomes of information on how PDC and CDMA transmission protocols work. I was happy to see that CDMA (Code Division, Multiple Access) used concepts that I learned in ECE 309, and that PDC was based on TDMA (Time Division, Multiple Access).I did quite a bit of C/C++ programming on my freetime.
Worked for Triton College as a lab consultant, and then as a UNIX system admin as soon as the college went "live" on the Internet. Fall 94 and Spring 95 were both semesters that I went heavily into MacOS programming and learning the AFP/DDP/MPP protocols of AppleTalk.
Worked at an OSCO drug store by my home.