Daniel Jarratt
Data scientist
I have 5 years experience with data science, software engineering, and product management, with a focus on recommender and decision support systems.
- I'm currently working on prediction of school dropout likelihood for state departments of education, where I can explore complex hierarchical time-based models.
- For fun I use R and Stan to model football team strength and predict game outcomes, and I'm working on rigorous sports analytics model evaluation methods.
- I also spend significant time with data science in service of college search and higher education enrollment management.
R, SQL, Java, Python
M.S. in computer science, University of Minnesota
Twin Cities, Minnesota, USA
Education
Master of science in computer science — University of Minnesota
- GroupLens research lab: advisors Joseph Konstan and Brent Hecht
- created location-based movie recommender using geographic information in collaborative filtering
- published “Where Far Can Be Close”: Finding Distant Neighbors In Recommender Systems at Proceedings of the Workshop on Location-Aware Recommendations co-located with the 9th ACM Conference on Recommender Systems ()
- recommender system evaluation research assistantship
- user interface design and evaluation teaching assistantship
Bachelor of arts — Valparaiso University
- Phi Beta Kappa, summa cum laude honors
- television/radio production, humanities
Sports analytics
College football rankings, college basketball rankings, and NFL rankings built with R and MCMC sampling
Work experience
Data scientist — Infinite Campus
- prediction of school dropout likelihood for all high school students in Kentucky
Software engineer — Infinite Campus
June 2015 - May 2016
- software engineer for a student information system with 7 million students
- development in Java and SQL
Recommender systems consultant — employer confidential
April 2015 - May 2015
- advised company (2nd-largest in their industry) on designing and evaluating a recommender system
- developed collaborative filtering and data mining prototypes
- reported to the CTO and head of engineering
Software engineering intern — Google Now
May 2014 - August 2014
- envisioned and developed a location-based recommendation feature, extending the Google Now service, with support from multiple teams
- front-end development in Django, JavaScript, JSON, HTML, and CSS
- back-end development in C++, Python, and Linux-based data center deployment
Technical product manager — PossibilityU
November 2011 - December 2013
- product leader: defined and prioritized the product roadmap and managed five engineers
- owned the recommender system, including data modeling, designing an item graph and determining and using its properties, and recommendation ranking
- won national recognition and competitive awards for my product
- invitation-only White House Education Datapalooza in 2012 and 2013; highlighted in 2013 keynote address
- Software and Information Industry Association top 10 in 2012 Innovation Incubator
- winner of College Knowledge Challenge funded by Facebook and the Bill and Melinda Gates Foundation
- Chronicle of Higher Education on me and PossibilityU: Netflix-Like Algorithm Drives New College-Finding Tool
- New England Cable News on PossibilityU: Maine Company Helps High School Seniors With College
Software side project full-time — College Recommender
January 2011 - November 2011
My work in higher education admissions led me to many families struggling to discover relevant and affordable universities. College Recommender was my solution: a system to infer preferences from a set of acceptable colleges, then recommend better (more affordable, better academic or cultural fit) options with similar features. This project required significant creativity and skepticism about then-current market wisdom; it was the first college recommender, followed four years later by others in the industry.
- entrepreneur, software engineer, and product manager
- envisioned and built a system to infer college search preferences and create personalized recommendations: collecting and interpreting the right data, database and algorithm design choices, user experience design, and evaluation along the way with students, families, and college admissions professionals
- designed system to handle difficult properties of this recommendation domain: no rating or consumption history, complex preference elicitation, and multiple stakeholders, among others
- developed in PHP and MySQL
- sold this software to PossibilityU, who hired me as technical product manager