I started PredictionIO with the mission to scale and simplify the development of machine learning technology. When Salesforce acquired the company in 2016, we were excited about the potential to reach a wider developer audience. Last summer, Salesforce donated the open-source edition of our product to the Apache Software Foundation (ASF) — it was unanimously accepted into the ASF Incubator program, which kicked off this open-source adventure. The team has spent the last 14 months working with mentors and building our user and developer community, which has provided invaluable feedback to help us reach this milestone.
The process of graduating from the Apache Incubator takes countless hours, drive, humility and collaboration with other companies and individual contributors. The goal of the incubator is to foster and grow projects under a rigorous governing method so ASF can provide “reliable and long-lived software products” that the developer community can trust. By involving a passionate community, ensuring a continuous flow of resources and guaranteeing a well-maintained product, the ASF board then votes to grant projects a top-level status — the highest status for an open-source project. By joining the ranks of other well-known top-level projects such as Apache Spark and Apache Hadoop, developers can be confident that the platform is well-supported and ready to use in production.
Our team that supports this project — and also supports Salesforce’s larger push around artificial intelligence with Salesforce Einstein — is more dedicated than ever to democratizing machine learning. Apache PredictionIO gives developers the entire stack needed to easily create end-to-end intelligent apps and deploy them at scale without having to cobble together underlying technologies. Developers can also download ready-to-use open source machine learning engines — such as recommendation, classification, regression, natural language processing, and many more — from PredictionIO’s template gallery. Building a recommendation engine would traditionally take a team of expert data scientists months to do, but with Apache PredictionIO, it’s now possible to do in a couple weeks with one or two engineers.
So what can Apache PredictionIO help you achieve? As an example, a small team of Salesforce engineers created Agenda Builder, a feature within the Dreamforce App, that uses machine learning to suggest a conference schedule for attendees. Apache PredictionIO helps the app use your behaviors and other attendees’ preferences to recommend what sessions you’ll want to attend. Be sure to try it out before heading to Dreamforce!
Of course, there are many more use cases all over the world from companies large and small. Check out some of the Dreamforce sessions listed below to understand what Apache PredictionIO is.
Thank you to the Einstein team, as well as our community and Apache Software Foundation that has been instrumental in this incubation process. We are excited to see where this project takes us next.
Monday, November 6, 10:00 AM - 10:40 AM
Moscone West, 2009
Luciano Straga, Salesforce Certified Professional, Santex
Mon, November 6, 13:00 – 13:40
Metreon AMC Theaters, City View Presidio
Thursday, November 9, 10:00 AM - 10:40 AM
Moscone West, 2006
Simon Chan, Senior Director, Product, Salesforce