2nd International Conference on Predictive APIs and Apps

PAPIs '15 will take place on 6 & 7 August 2015 in Sydney and online

Big Data can only deliver an impact when associated with intelligence that extracts knowledge from it. This is what Machine Learning does by automatically finding patterns in data and using them to make predictions. These predictions can then be integrated in real-world applications and at large scale thanks to predictive APIs (Application Programming Interfaces). PAPIs '15 brings together experts from all over the world who are defining the future of predictive technology and who are exploiting its opportunities.

PAPIs (which is short for "Predictive APIs") is the premier forum for the presentation of new machine learning APIs, techniques, architectures, and tools to build predictive applications. It brings together practitioners from industry, government and academia to present new developments and identify new needs and trends.

The upcoming conference, PAPIs '15, will feature talks on real-world use cases, lessons learnt by developers of predictive apps, tutorials on tools and APIs for building such apps, and a research track (see program below). It will also be the 1st time ever that leaders at Amazon Machine Learning, Microsoft Azure ML, Google Prediction API and BigML meet on the same stage. Last year, PAPIs '14 took place in Barcelona and attracted more than 200 people coming from 22 different countries.

The conference is co-organized by GCS Agile and Persontyle.


  • Bob Williamson, Leader of the ML group at NICTA (Keynote TBA) 

  • Brian Gawalt, Senior Data Scientist at Upwork, will show how they built a better predictive API with the Actor model of programming, to anticipate who's looking for work on the platform (previously known as Odesk/eLance)

  • Sharat Chikkerur, Senior Software Engineer at Microsoft, will give a behind-the-scenes look at Azure Machine Learning by presenting the anatomy of the service

  • James Montgomery, Lecturer at University of Tasmania, will present a specification for presenting learning algorithms as RESTful web resources, allowing for interoperability and composition of data processing tasks and predictive APIs

  • Alex Housley, CEO at Seldon, will share his story of transforming the platform from a “black box” predictive API to an open-source one

  • Nicolas Hohn, Director of Analytics at Guavus, will discuss data science and engineering challenges in building a scalable API for anomaly detection on network data streams

  • David Jones, Technical Director at Resolve Digital, who increased revenue of an online wine retailer by 71% with a recommendation API powered by small data and open source software PredictionIO

  • Yan Zhang, Data Scientist at Microsoft, will present a tutorial for building predictive maintenance applications, covering the industry landscape, how to formulate the problem in different ways, and how to compare approaches

  • Mark Reid, ML researcher at NICTA, will moderate a panel discussion on the research challenges facing predictive APIs, with leaders at Amazon ML, Microsoft Azure ML, Google Prediction API and BigML

  • And more coming soon... Sign up to get updates!


Practical information

Louis DorardComment