Modern data infrastructures for modern companies

Forward.
Forward.
Grupo público

Cabify

Calle de Pradillo, 42, 28002 · Madrid

Cómo encontrarnos

VOUCHER DESCUENTO 25% CABIFY: FORWARD.0503

Imagen del lugar del evento

Detalles

In this edition of Forward we will talk about Modern data infrastructures for modern companies.

Talk: Building a Data Lakehouse: Lessons learned from building modern data platforms for the Enterprise”

Marc Gonzalez, Freelance Data Engineer associated with InnoIT Consulting, is going to introduce the common components of a modern data platform in the Cloud, leveraging near real-time analytics solutions, provide some tips and tricks to avoid classic pitfalls and finish strong by showcasing how data processing frameworks like Apache Spark or Apache Beam can help in this endeavor.

Talk: Lykeion: Cabify’s Machine Learning as a Service

José Honrado, Data Engineer at Cabify, will talk us about how the life of a Data Scientist is not always easy. They can create fancy machine learning models to solve really complex problems. However, usually, that’s not the hardest part of the job: trouble comes when it’s time to see that model making decisions in a production environment.
Usually, that model was created in a language that is not part of the tech stack of the live services. So now what? The model must be rewritten and handed over to a product team to run it and maintain it. And then, the data scientist loses control of that model and gets in complete devastation and pain... At Cabify they take really seriously the well-being of their data scientists, so they've developed an MLaaS (Machine Learning as a Service) platform, for them to control the complete lifecycle of machine learning models and features: from training to production, and then... to evolution.

Talk: Making AI happen: Hard-won insights to navigate the AI software development lifecycle

Jesús Montes, Data Science & Data Engineering Lead at The Cocktail, will talk us about how developing a successful proof-of-concept in a data science project is a trigger of immediate sense of pride and enthusiasm for all the teams involved. However, this initial victory generates inflated expectations about achievable impact that, more often than not, are blown up when confronted with the harsh reality of taking a model into a production environment. He will share a story of mistakes, failures, regrets and occasional wins. The story of the journey to understand that data science and software development are, to a large extent, just the same thing.

After the talks, we have beers and pizzas! Also, we will have vegan and vegetarian pizzas!