Wide-eyed, outside of Dr. Hubert Dreyfus's door at Berkeley - 2007
I am a Masters student in Computer Science at Georgia Tech. I am also a Systems Analyst III in the Academic and Research Technology Directorate of the Office of Information Technology at Georgia Tech. I started the Masters program in January 2007 and my employment with GT in September 2006.
Prior to moving (back) to Atlanta, I spent four years as a Senior Software Engineer at Appian, a leading Business Process Management software firm.
I grew up in Atlanta, attending Sagamore Hills Elementary, Pace Academy, and Emory University before graduating Summa Cum Laude from the Robinson College of Business at Georgia State University in 2002 with a BBA in Computer Information Systems. I worked my way through GSU as a software developer (I had not yet earned the title Software Engineer. Despite lack of certification standards in the industry, I think that I can now claim to be an engineer, even, perhaps, a helluva engineer. I have, however, been a ramblin' wreck for years.).
I am currently exploring opportunities for PhD studies in various programs congruent with my research interests.
I am interested in the intersection of software engineering and cognitive science - specifically, the cognitive processes involved in performing software engineering. This is very broad, so let me illustrate with a few examples:
So much for my domain of interest. Techniques? I am still open to many approaches. Following the Dreyfus Model, I suspect that novice cognition (when the novice is instructed, thus "directed" skill acquisition) is largely rule-governed but that expert cognition is experience-based and even, perhaps, not explicable by rules (quick note: by "rules" I mean the rules of the symbolic approach - I am not denying that a connectionist, neurological, or dynamical systems approach may explain expert behavior - indeed, this is my hope and assumption). Thus I am looking for models that will shed light on both symbolic and non-symbolic cognition and explain how a single software engineer can move between both modes of reasoning, over both the time scale of a professional lifetime and within particular problem-solving episodes (a much shorter, but still cognitively long, duration). My current focus is Integrated Connectionist / Symbolic Architecture developed by Paul Smolensky and his collaborators and presented in the 2006, two-volume work The Harmonic Mind.
I work in the Design Intelligence Laboratory at Georgia Tech led by Dr. Ashok Goel, my advisor. We are investigating the automatic adaption of software agents based on models annotated with functional and teleological specifications. My current focus is on facilitating the translation of noun-based software agents into TMKL (our modeling language, which is biased towards verb-based approaches).
"Wait!" you say. "What does this have to do with your stated research interests?" At least, this has been some people's response. Well, let me explain. First, this work provides me with a good foundation in symbolic AI, which is not a bad place for a cognitive scientist's education to start. Second, model-based adaptation is an alternative view of learning, one that is applicable to purely symbolic architectures, and thus a potential modality of how novice software engineers learn. Three, down the road a bit, we will be building software tools and that will be a good time for me to develop my HCI skills, an interest I mentioned above. Even now, this project is driving me to research the relative affordances of procedural and object-oriented decompositions for engineers. Of course, I will approach this problem with the idea in mind that it is unlikely that novice and expert engineers are cognitively identical with respect to various modes of decomposition. I am convinced that I am well-situated to learn much from this work that is relevant to my overall project. And, as we all know, overall projects of grad students are subject to change, so I am open to that, too.
None yet, but watch this space, since I hope for this lack to change soon...
I wrote a term paper for CS 6795, Introduction to Cognitive Science, suggesting that the Dreyfus Model might be better explained by the Integrated Connectionist / Symbolic Architecture than by any other extant architecture. I have finished the class, but am still working on the paper. I will post it when I feel comfortable with releasing it to the world.
I wrote various papers as an undergraduate, a few of which are not completely sophomoric. I am still deciding whether any of them are worth posting.
I also write prolifically as a mode of thought, about whatever I am working on at the moment. Any of these working papers that I can polish but that are not going to be published are candidates for posting here, as well.
Basically, as with publications, stay tuned!
I just completed CS 6795, Introduction to Cognitive Science, with Dr. Ashok Goel. In Fall 2007, I plan to take CS 8893, Special Topics in Cognitive Science: Cognition and Culture, with Dr. Nancy J. Nersessian.
As an undergraduate at Georgia State University, I took two particularly relevant, graduate-level courses. One, Intelligent Systems: Development and Use with Dr. Balasubramaniam Ramesh of the Computer Information Systems Department. Two, Data Mining with Dr. Satish Nargundkar of the Decision Sciences department.
I am a member of the Cognitive Science Society.
I don't have a scholastic CV yet, and my resume isn't fully up-to-date, but here's my industry resume as of August 2006 (it's hosted in the open-source Vice project, which I run).
I am a LinkedIn personage, so I always welcome contacts through other people I know.
You may reach me at my first name + '.' + my last name at 'gatech' in the 'edu' TLD, if you think your mail will catch my fancy.
This document was hand-crafted in emacs on a MacBook Pro under OS X while in California on vacation. I'm a software engineer, not a web designer, so address any graphic design complaints to /dev/null.
May 2007