Improve Quorate's transcript readability and usability. Responsibilities:- Research and implement punctuation insertion (Python, TensorFlow)- Design and implement transcript search/analysis tools (Scala, Elasticsearch, CouchDB, RESTful API)- Develop web applications that demonstrate API utility (HTML, Javascript, Angular, D3)

Design and implement a tool for conducting social science research regarding the use of social media.Responsibilities:- Implement and evaluate a prototype system for clustering microtext by topic- Collaborate with potential users to define use cases and guide development - Data visualization

Responsibilities: - Implement a topic clustering system for high volume, streaming social media content to be applied as a tool for newspaper journalists and editors - Produce scalable, efficient, readable code with documentation and test scripts - Provide research results (published or own experimental) to support design decisions - Find optimal solutions to the unique challeges of clustering streaming micro-text, including: text normalization, new story detection, content selection and topic drift - Communicate and coordinate with colleagues throughout the UK to integrate the clustering system into a larger graphical work bench

Propose improvements to a text classification system based on extensive experimentation, on real world data, using the Weka machine learning toolkit. Instruct the AI development team on the use of Weka for the following: - data visualization - text classification - feature selection - dimensionality reduction

Work as part of a small team of computational linguists to ensure functionality of the Swype virtual keyboard in a wide variety of natural languages. Responsibilities:- Identification of language specific requirements- Keyboard localization including design and implementation- Standardization and documentation of keyboard design and implementation procedures- Training and managing linguistics interns- Training keyboard technicians- Obtaining and incorporating feedback from native language users/testers- Communication with external graphic designers and translators- Monitoring web crawler- Receptionist duties when necessary

Regular expression based processing of wordlists for entry into the PanLex pan-linguistic dictionary.