Applied AI Meetup #5: We're back!

We know it's been a long time since we had a meetup and a lot has happened - both in the world and at Ravelin. Nonetheless, we're keen to get the leading AI innovators up on stage again to showcase their work. We're headed back to Google Campus London, where we will hear from Guillermo Christen from Logameal, Katya from Codec and Ravelin's very own Fabian Peters.

Programme:

6:30pm - Beer, food and chat
7:00pm - Guillermo Christen
7:30pm - Ekaterina Volkova-Volkmar
8:00pm - Fabian Peters

What I Learned From Looking at a Million Pictures of Food

Guillermo Christen

Improving your diet starts with becoming aware of what you eat, and that is a lot harder than it seems. I will present a deep dive into why it's so difficult to teach a machine to recognise food, and why you would even want to do it.

Guillermo Christen was co-founder and CEO of Logameal (acquired Jan 2017), an app that helped people develop and maintain healthier eating habits by making it easy to track what they eat – just by taking a photo of their meals. Before that he led product and technology at two machine learning startups. TV Genius (acquired in 2011) and Intent HQ where he is currently CTO.

Supporting Content Creation with Machine Learning

Ekaterina Volkova-Volkmar

Codec helps brands remain relevant to their target audiences. Our content intelligence platform combines big social data with versatile machine learning solutions to tell brands what content resonates with the audiences they want to target, all before the brand spends big on high-production content. The result is interesting content, engaged consumers and increased marketing ROI.

In this presentation I'll cover what data methods can help to inform a creative content decision.

Ekaterina Volkova-Volkmar is a lead data scientist at Codec. Prior to this role she was a data scientist in the field of digital health. She graduated in Computational Linguistics and did a PhD in Neuroscience in Tübingen, Germany.

Shipping with Confidence - Lessons Learned from 1.5 Years of Deploying Models

Fabian Peters

At Ravelin, we solve fraud for clients with different data models across a number of industries. At any given time we have multiple machine learning models in production and we have to ensure that changes we introduce do not negatively affect performance for any one of them. This talk is about how we make sure that our models are reliable and safe to deploy whilst minimizing the amount of manual work. I focus on five lessons which go beyond classic measures of performance.

Fabian is a data scientist at Ravelin where he fights fraud using machine learning. Before joining, he studied the changes in energy consumption of real neurons at Oxford and UCL. He has a PhD in Neuroscience.