My name is Devon Terrell. I’m a software developer with a background in mathematics and data science. My primary interest is language.

Language is a part of almost everything we do as humans. It allows us to communicate our ideas to one another. From a humble conversation between a cashier and customer at a coffee shop to great written works from antiquity, natural languages form the core of human interaction. Increasingly, digital communication is becoming not only common but the standard for modern communication, building ever higher mountains of data to be analyzed.

Mathematical notation is a language many struggling students consider foreign. The rich and constantly evolving language of music concisely represents concepts that would otherwise be cumbersome for performers to interpret in real-time. Computer programming languages and communication protocols are rigid languages, following precise, formal rules. Even within natural languages, fields such as medicine are built around highly technical, domain specific sub-languages.

A key step in solving complex problems is to build up a language to describe the problem and frame potential solutions. Whether verbal or graphical, this language is both shaped by and further informs the mental model of those that use it. Properly applied, language allows us to break down complex problems, to abstract away burdensome details and focus on the essential aspects. Poorly formed, language limits expressivity, leading to misunderstanding and ambiguity.

Natural languages processing techniques, compilers for computer programming languages, and even entire grammars enabling users to build interactive data visualizations all have roots in understanding and interpreting language.

Technologies

Education

• B.S. Mathematics at Virginia Tech
  • Minor in Computer Science
• Certification from Cloudera as Certified Cloudera Hadoop Developer
• M.CS with a focus in Data Science from University of Illinois Urbana-Champaign
  • Partial Completion

Work

• 2018 - Present Textio

  • Built high-speed middleware processing system for augmenting event data
  • Implemented graph-based service for querying company metadata, including supporting sourcing and time-based queries
  • Implemented several information retrieval and search services as part of larger teams for integrating with external data sources
  • Worked with Data Science teams to refactor analyses producing critical business metrics into scalable, scheduled jobs which integrate with Salesforce.

• 2015 - 2018 CareEvolution

  • Designed and implemented a drag-and-drop editor to allow users to build custom forms and reports with dynamic data binding
  • Built UI for healthcare term disambiguation system
  • Built UI for user administration system

• 2010 - 2015 Sphere of Influence Inc

  • Designed and implemented proof of concept data architecture for exploratory analysis against multi-source data system
  • Worked closely with data science team to create scalable models for geospatial, agricultural, weather, and telemetric data
  • Developed UI and visualization tools for distributed network analysis and cyber security tool
  • Designed and developed R&D named entity recognition and disambiguation tool

• 2008 - 2009 U.S. Department of Defense

  • Performed forensic examination of various digital media
  • Built custom scripts to scrape data from digital forensic images and generate reports