AWS IoT in Practice: From Core Concepts to Hands-on Implementation

Session Abstract
This four-hour workshop will introduce participants to Amazon Web Services’ IoT ecosystem, with a specific focus on AWS IoT Core and AWS IoT Greengrass. Attendees will begin with a foundational understanding of AWS IoT services and their real-world applications, before diving deeper via a hands-on session with AWS IoT Greengrass. The workshop opens with a thorough overview of AWS IoT architecture and capabilities, then transitions into an immersive hands-on experience where participants will configure Greengrass devices, develop and deploy components, implement inter-process communication, and work with local MQTT brokers. Through guided exercises, attendees will gain practical experience in building resilient IoT solutions that can process data locally, respond intelligently to local events, and securely communicate with the cloud. This session is ideal for IoT developers, solutions architects, and technical professionals seeking to enhance their practical knowledge of edge computing and IoT implementation using AWS services. Participants should have basic familiarity with AWS and IoT concepts. Attendees will need to bring their own laptops with administrator access for the hands-on portion of the workshop.

Download full agenda here!

Darren  Roback
Darren is a Sr. Solutions Architect with Amazon Web Services based in St. Louis, Missouri. He has 20 years of experience in IT and is passionate about Data Analytics, Generative AI, Internet of Things (IoT), and Security and Compliance. At AWS, Darren partners with energy and utility customers to help them solve business challenges with AWS technology. Outside of work, he enjoys woodworking and spending time with his family.

Machine Learning and AI on AWS: Amazon SageMaker AI

Session Abstract
The Amazon SageMaker Immersion Day is a comprehensive hands-on workshop designed to provide participants with end-to-end understanding of building machine learning use cases using Amazon SageMaker AI. This session guides attendees through the complete ML lifecycle, from feature engineering and algorithm selection to training, tuning, and deploying ML models in production-like scenarios.

Throughout the workshop, participants will explore key concepts including feature engineering, model deployment, and an introduction to MLOps practices. The session offers flexibility in the second half, allowing attendees to choose between exploring Bring Your Own Model (BYOM) capabilities or diving deeper into MLOps with SageMaker Pipelines. Participants will gain hands- on experience with the SageMaker Console and Jupyter Notebooks, working with sample datasets to perform feature engineering and experimenting with Amazon SageMaker's built-in algorithms, particularly XGBoost, to build optimized models.

Download full agenda here!

Archit Malpure
Archit Malpure is a Solutions Architect at AWS supporting State and Local Government and Higher Education. With 5 years of AWS experience, he specializes as a cloud infrastructure generalist who works backwards from customer business outcomes. Archit focuses on designing and delivering solutions that align with the unique requirements of public sector organizations, helping them implement AWS best practices to achieve their strategic goals.

Kai Dickman
Kai Dickman is a Senior AI and Machine Learning Specialist at AWS supporting Public Sector customers. With over 25 years of experience spanning on-premises and cloud technologies, Kai works with both executive stakeholders and technical teams, architecting scalable machine learning solutions and implementing responsible AI best practices. In addition to a passion for all things AI, Kai enjoys connecting the real world to the Cloud with IoT.

Federated AI with Flower
Abstract 

Federated AI offers a promising path toward building secure, collaborative, and adaptive systems without the need to centralize sensitive data. In this tutorial, we introduce the Flower framework and showcase its use in scenarios such as anomaly detection, intrusion defense, and scalable federated deployments. Through hands-on examples, participants will gain practical insights into designing resilient Federated AI solutions while also reflecting on open challenges and future research directions for trustworthy and secure machine learning.

Facilitator Bios:
William Lindskog-Münzing: William is a Solutions Engineer at Flower Labs, with industry experience in the heavy-asset industries. His research experience at TU Munich includes federated learning for tabular data, and machine learning application in the automotive industry. He recently led the research and development team at TUM.ai, Germany's leading student AI lab. William also contributed with FedPer baseline during Flower's 2023 Summer of Reproducibility initiative.

Dimitris Stripelis: Dimitris Stripelis is a Research Engineer at Flower Labs, with industry experience at Amazon AWS, and Salesforce. He earned his PhD from the University of Southern California and worked as a Postdoctoral Researcher at USC's Information Sciences Institute (USC-ISI). He is a recipient of the USC Myronis Fellowship (2020) and the A.G. Leventis Foundation Educational Grant (2019–2021). His research focuses on federated and distributed machine learning. He has served as a reviewer for many conferences, including NeurIPS, ICML, AISTATS, AAAI, EMNLP, ECAI, and WISE, as well as journals such as TNNLS, TMI, and TKDE. He co-organized the first Federated Learning Systems (FLSys) workshop at MLSys 2023 and the Federated Learning on the Edge (FLEDGE) Symposium at AAAI 2024.