Building Reddit's Custom Time On Site Metrics with Airflow and Google BigQuery

Are you going?

74 people going

Share:

General Assembly San Francisco

225 Bush St · San Francisco, CA

How to find us

Please note it is the EAST lobby entrance

Location image of event venue

Details

Please be sure to RSVP both on our Meetup and the GA link here: https://generalassemb.ly/education/building-reddits-custom-time-on-site-metrics-with-airflow-and-google-bigquery/san-francisco/87784

Tentative Schedule:
6:30pm: Pizza + Beer networking
7:00pm: Bad Boys, Whatcha Gonna Do: Predicting Crime on the Streets of SF with Ruqaiya Shipchandler, Solutions Engineer at Dataiku
7:30pm: Building Time On Site at Reddit with Katie Bauer, Data Science Manager at Reddit

Talk Abstracts:
Building Time On Site at Reddit with Katie Bauer, Data Science Manager at Reddit:
Time on site is a foundational metric in web analytics and building it seems straightforward enough. But modern websites are built on the backs of distributed systems, and distributed systems make it particularly difficult to figure out when something actually happened. In this talk, we'll discuss how we implemented our own time on site metric, building ETLs with Google BigQuery and Apache Airflow, as well as the choices we made to do it, the problems we caused with those choices, and how we fixed them.

Bad Boys, Whatcha Gonna Do: Predicting Crime on the Streets of SF
with Ruqaiya Shipchandler, Solutions Engineer at Dataiku
While San Francisco is most famous for being the technological epicenter of the world, the city's infamous past as the home of notorious criminals at Alcatraz makes us wonder: what is SF's current criminal landscape? And can we use data science to proactively fight crime in the city?
We'll share how we used Dataiku DSS to explore over 12 years of SF crime data to understand key trends, and build a predictive model to pinpoint the category of crime that would occur given time and location.

Speaker bios:
Katie Bauer was a founding member of Reddit's data science team and currently manages its Consumer Data Science and Analytics team. She has previously worked in search, digital advertising, and online retail, and is known for using company hackathons as an excuse to bake cakes at work.

Ruqaiya is a Solutions Engineer at Dataiku based in the East Bay. She works with companies to address their data science challenges and implement efficient, sustainable solutions. Ruqaiya’s initial exposure to data science was through her work in the Energy industry, where she had the opportunity to develop predictive asset maintenance models, models to minimize environmental impact, and improve employee safety. Ruqaiya has a degree in Chemical Engineering from the University of Houston.