Thursday 1:30 p.m.–3 p.m. in Room 13

Bridging the gaps between Science and Engineering with Jupyter Notebooks

David Taieb


When it comes to data science in the enterprise, organizations often face a culture of self-interest that leads to unnecessary friction between the Data Science and Software engineering teams, causing inefficiencies, and ultimately a failure to realize the full potential of data science. In this talk, we’ll show how Jupyter Notebooks combined with the PixieDust open source library can address the expectations and challenges around data science, including the shortage of technical expertise, overly complex and inaccessible analytics tools and more.


Jupyter Notebooks are very popular with data scientists because they help them import, visualize and explore data at scale, all within a single environment. PixieDust not only accelerates these steps with an interactive API to build powerful Data Visualization, it also provides an elegant Python-based programming model called PixieApp. PixieApp enables developers to collaborate with Data Scientists within the Notebook to build powerful dashboards that combine a sequence of analytics into one compelling interface and then deploy them as stand-alone web apps using the PixieGateway micro-service server. We’ll finish the session with an in-depth demo, including code walkthrough, of a Python Notebook that analyzes Stocks financial data and then operationalizes its analytics by building and deploying a real-time dashboard.