2807 NW 25TH ST

April 3, 2017

Lawton, OK 73505, U.S.A.

To Whom It May Concern:

I was a programmer for Alatec Inc., at the Fires Battle Lab in Fort Sill, OK in the modeling and simulations section. I programmed in a C++/Linux environment. I passed my 90-day review on June 1, 2013. This battle lab is the best one in the entire Army. My tasks were basically scientific programming, and unit and system testing of the code I produced. I was also using GTK+2.0 for help in the creation of a new Linux user GUI for FIRESIM XXI.

My former team supervisor at CGI, Ron Duran, was very jealous about my new job at the Battle Lab. He wanted me to lose that job, and did all he could to make that happen. He was a former employee of the Battle Lab, and was still very involved with the Lab, although he should not have been. He used his influence there to convince them to cause me to lose my job. The project manager and several others there lied or committed acts that were highly unprofessional, and I did end up losing my job. I did complain to the Army Attorney General. I do know that Ron Duran was forced to quit by them, due to his illegal abuse of narcotics, and perhaps other reasons. I never received any information from the Army Attorney General, my statement above about Ron Duran was a deduction based upon my report to them and whatever I had heard afterwards when applying to CACI, who took over the Lawton CGI building.

I was a Software Engineer/Contractor for the Army at CGI in Lawton. I had experience with the profiler for approximately 3.6 years, and experience with radar for approximately 1 year. I spent 3 months designing and developing the code to solve a business-related problem for the section managers and upper managers. I came up with the inspired algorithm in several hours, one evening.

I also have 12 years of experience in Heart Rate Variability (HRV) research. During this time, I primarily co-authored or authored 41 journal articles, 34 abstracts, and 4 presentations. I gained 12 years of programming experience during this time, and a significant fraction of this experience was programming advanced mathematics. I gained about 6 years of experience programming advanced mathematics during my Post Doctorate and Doctorate research.

Due to funding problems, my HRV position in St. Louis ended April 18, 2008. I have a letter from the department of Cardiology at Washington University School of Medicine stating this fact. I did not have time to finish my study with heart transplant patients, which was no longer being funded.

Please do not contact Phyllis Stein, Ph.D., my former supervisor in Heart Rate Variability. She will not provide an honest review of my performance. She always said I was her secret weapon, which meant that my skills and accomplishments were to appear as her skills and accomplishments.

By working at CGI, I have gained new experience by analyzing very large software systems. I learned that I am very strong in code analysis. I have found many errors in the logic of the code I have analyzed. In addition, I have improved my UNIX, Linux, Java, shell scripting, and software engineering skills. Light experience in electrical engineering and meteorology has also been gained.

The discovery of the bug in the qrsdk.c program was the result of an inspiration. The programming and design of the large HRV analysis program was a good mental exercise, since it required that I hold the entire program in my mind while working on it. I always read input data using random access methods, and I used dynamic memory allocation as much as possible. The program required scientific programming, character string manipulations, and the handling of many files and variables.

Upgrading the HrvInteractive program to use MSVS2005Pro was relatively straightforward. Upgrading HrvInteractive to us MSVS2003Pro required re-programming some of the dialogue boxes and other graphics. The final additions to the program dealt with adding tachogram displays and 24-hour HRV plots to aid in obtaining bed-wake times. The program worked very well, skipping from one time segment to another in the blink of an eye. The usual time segment for HRV was 5 minutes. The in-bed/out-of-bed HRV displays were presented more slowly since many hours of HRV data were presented on 4 different panes.

Computerized automation of manual tasks involved the implementation of new shell scripts or C programs involving many system calls. Often, Cygwin was utilized as the OS for running the automation programs/scripts when the OS was windows. When a UNIX OS is indicated, Solaris was almost always utilized, as opposed to a Linux OS.

The custom output of HRV variables into a *.csv file, which is automatically read by MS Excel, required additional programming of the HRV code written by J. Rottman, M.D. When the large HRV program mentioned above was completed, only a file had to be edited in order to select the desired HRV variables for output into a *.csv file.

Polysomnograms were usually in EDF format, and contained between 11 and 30 signals. I wrote code to be able to plot some of these signals in the same format as the heart rate tachogram I developed. These were hourly plots, and they were very helpful for a qualitative analysis. The most interesting plots were those of the sleep stages over the entire night. The heart rate tachogram plots, which were usually made by scanning the ECG signals with a Holter scanner, were a useful aide in detecting sleep apnea.

I personally took care of the IT operations of our HRV lab alone, for a significant part of my 12 years there. I did manage some students and helped manage their projects over that period of time. I was the part-time UNIX and Linux system administrator for 8 years in the HRV lab. I managed one RHL 7.3 server, one Solaris 7 server, and two Solaris 8 servers. The servers were an IBM PC, a Sun SPARCstation 5, a Sun Ultra Enterprise 450, and a Sun Ultra 60, respectively. I also spent years administrating the x86 PC’s (WINXP Pro, WIN2000 Pro, WIN98) in the lab.

Sincerely yours,

Peter P. Domitrovich, Jr., Ph.D.