Jeremy T. Bakker, PhD

Biography

I am a software developer drawn to the efficient interpretation and clear presentation of complex data sets. For over a decade, I studied and worked in the highest echelons of the academic world, honing my critical thinking skills among some of the world's best thinkers. All the while, I never forgot my roots. I was a latchkey kid as early as six years old and spent years working among youth involved in inner-city street gangs.

Software development is, in many ways, my true home. I have long been fascinated with computer programming, but I never pursued it seriously because I believed that I could better reach my potential through other pursuits. After completing my PhD and working alongside several technology entrepreneurs in Nashville to revitalize a historic non-profit organization, I came to realize that my skills are indeed better suited for software development. I am a learner, a leader, a creator, and a collaborator with an unparalleled ability to digest a large amount of complicated information quickly and to communicate it concisely to diverse constituencies. Having worked for years in international organizations committed to abstract ideals, I am ready to make a tangible contribution to my local community and its institutions.

Here you will find some of the ideas and organizations that have shaped me. Take a look at my influences and trace my recent path. See my resume for a lengthier and more formal documentation of my skills and professional journey. Or check out some of my projects below.

Lem(m)a

Lem(m)a is a scalable, single-page JavaScript application for natural-language processing built in the Angular framework. It measures cosine similarity between the tf-idf vectors of a query and a control set of data. Because it incorporates Browserify, it can use all the algorithms of the "Natural" node module. Nonetheless, I coded Lem(m)a's tf-idf algorithm myself because I wanted to know more about its strengths and weaknesses.

Lem(m)a is my first step in testing a theory I have about the use of digital tools to quantify qualitative human phenomena. After reading Brené Brown's Daring Greatly, I decided to test the limits of my ability to evaluate abstract ideas. Early in that book, Brown writes, "I was riveted by a statement from one of my research professors, 'If you can't measure it, it doesn't exist'"(8). I wanted to know how far I could take that idea, so I began with a topic for which I have expertise: religious struggle.

Lem(m)a is the ancient Greek word for "theory" or "proposition," the koine Greek transliteration of the biblical Hebrew word for "why," and the technical term for the dictionary entry of a word. I have a theory about the use of ancient Israelite psalms of lament to quantify religious struggle. This application parses the data for me and presents it in an intelligible form with the help of D3.

Check it out on GitHub!

Try it here.

Prometheus

Prometheus is a Python application that uses the Django framework, the Natural Language Toolkit, and a SQLite database to correlate corporate earnings reports and linguistic data from quarterly earnings call transcripts. I built the tool to test social psychological and linguistic theories on detecting deception in discourse. Inspired by James W. Pennebaker's The Secret Life of Pronouns: What Our Words Say About Us, I want to determine what linguistic data can predict about a company's value in the US stock market following quarterly conference calls. Prometheus quantifies CEO and CFO patterns of speech in concert with its display of earnings data over time.

Check it out on GitHub!