Airflow Adventures: Streamlining Demand Forecasting & Operationalizing LLM Apps"

After a long pandemic-induced hibernation the Data Council Amsterdam Meetup group is back! Join us for some tales of adventures with Apache Airflow! We’re delighted to welcome 2 excellent speakers to share how they are using Airflow to push the limits in very different parts of the data landscape.

📅 Agenda:
17:45 - 18:30h 🍕 Walk-in + Pizza & Drinks
18:30 - 20:00h 🎙️ 1st talk, short break, 🎙️2nd talk
20:00 - 20:30h ☕️ Networking

📍 Location:
Xebia Data
Wibautstraat 202, Amsterdam

Sign-up instructions:

• Sign up via meetup.com
• Let us know if you have any strict dietary restrictions (e.g. vegan🌱)
• We are looking for speakers for the next events. If you would like to give a talk, let us know the topic and a contact information.

🎤 Topics:
👉 1st talk: Streamlining Demand Forecasting at Scale
By Rogier van der Geer (Xebia Data) & Anton van Steenvoorden (Albert Heijn)

Leveraging Airflow for Forecasting at Albert Heijn.
Join us as we explore how Albert Heijn, the leading supermarket chain in The Netherlands, leverages Airflow to streamline Demand Forecasting. With the need for over half a billion daily predictions across 1k+ stores and 15k+ products, the challenges are immense. Discover how Airflow's orchestration capabilities enable efficient training and running of over 2000 models, ensuring accurate and timely forecasts. Our focus will be on Airflow, and how it enables the training and running of thousands of models. We will delve into the intricacies of automatically generating the workflows, ensuring seamless execution, and maximizing the visibility of any issues. From data ingestion to model training and deployment, we will discuss the strategies and best practices employed to achieve accurate and timely forecasts.

👉 2nd talk: Operationalizing LLM Apps
By Michael Gregory (Astronomer)

Abstract: This session will present tools, processes and frameworks for taking LLM applications from prototypes to production with Apache Airflow. We will present Ask Astro a production application and open-source reference implementation for Andreessen Horowitz’s Emerging Architectures for LLM Applications.

Description: Retrieval Augmented Generation (RAG) with large language models (LLMs) is creating new opportunities for enterprises to leverage their institutional knowledge to provide better customer experiences, new products and simplified operations. While building RAG-based application prototypes is simpler than ever before, operationalizing these applications is still a challenge for many organizations. This presentation will discuss best practices and introduce open-source templates to help enterprise data teams operationalize RAG-based applications with Apache Airflow.