13:00 - 13:45

SERVERLESS-01

Event-Driven and serverless in world of IoT

Room: Yasmin-I | Time: 13:00

In large-scale IoT systems, where thousands of devices are constantly communicating, data and control messages can flow unpredictably and at varying rates throughout the day. Serverless and event-driven architectures provide a dynamic and efficient solution to this challenge. In this session, we’ll explore why serverless is a great fit for IoT, particularly in handling data ingestion and management. We’ll dive into the key architectural patterns used in IoT systems, highlighting how event-driven design enables scalability, flexibility, and cost optimization. I’ll also walk you through an architecture I designed for an IoT use case, showcasing the integration of various AWS services tailored for data ingestion, processing, and analytics. Along the way, we’ll discuss the challenges encountered and lessons learned during development, offering actionable insights into the unique aspects of building serverless IoT solutions. Join me to understand how serverless and event-driven architectures can unlock the full potential of IoT, enabling resilient and scalable systems that are easy to manage and adapt to evolving demands.

Jimmy Dahlqvist 🇸🇪

AWS Serverless Hero, AWS Ambassador, AWS Certification Lead SME & Cloud architect

AI&DATA-03

You don’t think about your Streamlit app optimization until you deploy it to AWS

Room: Yasmin-II | Time: 13:00

Building Streamlit apps is easy for Data Scientists but when it’s time to deploy them to the cloud, challenges like slow model loading, scalability, and security can become major hurdles. This talk bridges two perspectives: the Data Scientist who builds the app and the MLOps engineer who deploys it. We'll dive into optimizing model loading from Hugging Face Hub, implementing features like autoscaling and authentication, and securing your app against potential threats. By the end of this talk, you’ll be ready to design Streamlit apps that are functional and deployment-ready for the AWS.

Darya Petrashka 🇵🇱

AWS Community Builder, AWS Authorized Instructor & Senior Data Scientist @SLB

AI&DATA-05

AWS DeepDoc Scan

Room: Yasmin-III | Time: 13:00

DeepDoc Scanner is an AI-powered solution designed for searching structured and unstructured data efficiently. It automates document processing, eliminating manual work and minimizing the risk of errors. The system enables fast and precise retrieval of critical information, improving access to essential business insights. With built-in auditability and traceability, it enhances operational efficiency and compliance. DeepDoc Scanner supports digital transformation by converting paper-based documents into digital formats. It facilitates data categorization and analysis across multiple formats, including PDFs, Word, Excel, and OCR-processed images. The solution seamlessly integrates with existing data sources, ensuring smooth connectivity and accessibility. It features permission-based content filtering, allowing users to access only authorized information. DeepDoc Scanner includes various connectors and identity management capabilities to support secure data processing, making it a reliable and efficient tool for modern businesses.

Roman Čerešňák 🇸🇰

AWS Solutions Architect @DevOpsGroup