My first real job (1996 – 2000) was with the former Environmental Research Institute of Michigan (ERIM). That company is no more, having been consumed by Veridian and subsequently extruded into General Dynamics. My work was in algorithm development, helping to distinguish between tanks and trucks in various sorts of sensor data, mostly radar and infrared. The cocktail party version is that I taught missiles which targets to blow up. I also worked a little bit on detecting landmines and unexploded ordinance. The majority of the work was software development in C, C++, PERL, and some truly horrible application development frameworks – including Khoros.
I had some great mentors and leaders while working at ERIM. They allowed me great latitude to think creatively about the problems we were facing. In one memorable instance, I was allowed a year of my own time to develop software standards and modules in an attempt to unify algorithm development across an unruly alliance of 12 R&D contractors working for the Army Research Labs in Adelphi Maryland. I spent two memorable summers onsite at ARL, throwing myself at various problems and being thrilled at the per-diem meal pay I was receiving.
In 2000, I moved to Minnesota, and started working with a bioinformatics group – the Center for Computational Genomics and Bioinformatics. I was with the CCGB from 2000 – 2004, developing software and building compute resources for a variety of life sciences researchers across the University. It quickly became obvious to me that a developer with no understanding of the underlying science is useless, particularly in a research organization. Therefore, I took advantage of the free tuition to take a variety of graduate level biology courses. Through a combination of aptitude and willingness, I took responsibility for more than a few servers, clusters, databases, and web pages.
We worked primarily with legumes, doing expressed sequence tag (EST) analysis. My largest single contribution was to build a wide area compute farm that stitched together the resources of four different compute clusters for running BLAST and other analyses.
In 2004, I relocated to Providence, Rhode Island. At the same time I accepted a position as the first employee with Bioteam, Inc. We are fundamentally a consulting group specialized in “bridging the gap” between life scientists and high performance computing. I led support for our Inquiry product, as well as developing the web services (SOAP and WSDL) interface to it. Presently, my consulting practice focuses on large compute clusters, large scale data storage, and infrastructure needs for next generation DNA sequencing. I direct all consulting activities within the company, and also serve on its board of directors.