Machine Learning for Pentaho Data Integration

Pedro Vale will talk about machine learning in Pentaho Data Integration
Pedro Vale will talk about machine learning in PDI

Machine learning is transforming the ways we live and work. At Pentaho Community Meeting, Pedro Vale will present plugins that help to leverage the power of machine learning in Pentaho Data Integration. I have talked to Pedro about his talk and his job as Head of Development at Pentaho.

Who are you?
My name is Pedro Vale and I work at Pentaho Engineering helping to deliver the next versions of the Pentaho platform. I’ve been involved with Pentaho (and business intelligence) for the past 6 years when I joined Webdetails as Head of Development focusing mainly on CTools. When Pentaho acquired Webdetails we started working as part of the broad engineering group at Pentaho. That led to the growth of a strong Pentaho engineering team here in Portugal which I currently lead.

What is your connection to Pentaho?
I manage non-US engineering for Pentaho. This includes our engineering team in Portugal where we have about 40 people, our near-shoring team from EPAM based in Belarus and Russia and some other folks here and there.

What will your talk be about?
I’ll be presenting some PDI plugins related to machine learning. The common goal for those plugins is to make it easier to use some machine learning toolboxes or particular algorithms from Pentaho Data Integration. For instance, one of them allows you to use Recurrent Neural Networks (DeepLearning4J) in PDI. Our plan is to make these available in the Pentaho Marketplace so that community users can leverage them while building their projects, provide feedback and use them as examples for other related plugins.

The plugins were developed in a particular way – can you say more about it?
For the past three years now, we are running a couple of summer internships every year here in Portugal. We collaborate with one of the main technical universities here (Instituto Superior Técnico) and we provide students in their final year with some exposure to a work environment. We usually focus these internships on 1) items not on our near-future roadmap and 2) deliverables that can be either integrated into the product at some point or made available for others to use. These are short internships lasting usually a couple of months, so some of the work might be very specific. But we’ve been having really good outcomes, students grab the opportunity and really run with it, which by itself is rewarding. As a side bonus, these internships also help us to identify talents that we can later recruit.

What do you expect from PCM?
The usual mix of really interesting talks about innovative uses of the Pentaho platform, meeting folks we interact with every other day, dinner and drinks with the community people after the sessions. I’m also looking forward to the wine tasting Jens is setting up.