DAY 2 | FREE TRACK
Embedded Systems in Action: Building Smart, Resilient IoT Devices
10:00 - 10:15
Chairperson’s welcome
Chairpersons welcome and opening remarks.
10:15 - 10:45
Presentation: The Future of Embedded Systems in a Hyperconnected World
Embedded systems are evolving fast. As IoT networks grow, we’re seeing a shift in architectures, tighter AI integration, and major leaps in connectivity. Devices are getting smarter, more autonomous, and more deeply woven into our digital infrastructure. This session looks ahead at where things are going and how embedded systems are quietly powering the next phase of hyperconnected tech.
10:50 - 11:10
Presentation: Agentic AI Meets Embedded: How AI Agents Are Transforming Embedded Systems Engineering
Embedded systems engineering has long been defined by tight hardware constraints, manual debug cycles, and slow iteration loops. But a new wave of AI agents is changing how we write, test, and deploy embedded software from the IDE to the vehicle.
This session draws from real-world experience building and deploying agentic AI tools for embedded development,including: – Agentic IDE for microcontrollers: How Dhruvil forked the Arduino IDE into a Cursor-style agentic environment where AI autonomously writes code, compiles, flashes firmware to hardware, reads serial output, and self-corrects — iterating up to 12 rounds without human intervention
– AI on the edge: Lessons from building an ESP32-based AI robot with real-time voice interaction powered by GPT-4o — bringing agentic capabilities directly onto resource-constrained embedded hardware
– Agents in the autonomous driving pipeline: Perspectives from 8+ years integrating perception, sensor fusion, and controls software on NVIDIA Drive platforms at companies like NVIDIA, Lucid Motors, and Canoo — and where agentic workflows can accelerate development, debugging, and deployment
– What actually works today: Practical patterns for connecting LLM agents to hardware toolchains (compilers, flashers, serial monitors, JTAG) and the failure modes you’ll hit along the way
Whether you’re building IoT devices, automotive systems, or robotics, this talk will show you how agentic AI can collapse your embedded development cycle from days to minutes — with live demos.
11:15 - 11:35
Presentation: Securing the Intelligent Edge: Why Kernel-Level Protection Is the Answer to CRA Compliance
The EU Cyber Resilience Act is reshaping what it means to ship a connected product. For manufacturers of embedded IoT devices, it introduces mandatory security requirements that span the entire product lifecycle — from design through deployment and beyond. But meeting those requirements demands more than documentation and patching cycles. Traditional security approaches built for perimeter defense or cloud-based detection simply don’t scale to devices with constrained resources, no persistent network connection, and real-time operational demands. This session explores a new security model designed specifically for the intelligent edge: one that enforces policy directly at the OS kernel — blocking malicious behavior before it can execute, regardless of whether the exploit has ever been seen before. Attendees will walk away with a clear picture of what embedded-native, enforcement-first security looks like in practice, and how it simplifies the path to CRA requirements through vulnerability handling, secure defaults, and ongoing protection of devices in the field.
11:40 - 12:00
Presentation: Rethinking Device Manufacturing with SGP.32: Post-Issuance Customization at Scale
Industrial IoT manufacturers often struggle with customizing firmware, accessories, and configurations across diverse markets—tasks that still require physical access during production. With SGP.32 IoT eSIMs and built-in bootstrap connectivity, devices become digital from the moment they are powered on. This enables remote, API-driven customization of firmware, SIM profiles, and device settings after manufacturing, without locking in a connectivity provider. The result: faster scaling, less complexity, and a more flexible production process.
12:00 - 12:30
Networking Break
12:30 - 13:05
Fireside Chat: Navigating IoT Connectivity & Protocols
Industry experts explore the diverse and often complex world of IoT connectivity and communication protocols. They provide guidance on selecting the best technologies to match specific application requirements, weighing factors like power consumption, range, data throughput, and interoperability. The session also highlights how to navigate trade-offs and integrate multiple protocols within increasingly heterogeneous IoT environments to build reliable, efficient networks.
13:10 - 13:30
Presentation: Trust on the Edge: The New Frontline in IoT Cybersecurity
Join industry veteran Bill Lattin for a special session celebrating the recent release of his new book Trust on the Edge. A must-attend session for IoT manufacturers navigating today’s rapidly evolving cybersecurity landscape, Bill draws on decades of real-world experience securing safety-critical systems from silicon to cloud, Bill will share hard-earned lessons, practical strategies, and a blueprint for building lifecycle trust into connected products from day one. Attend to gain actionable insights—and grab a complimentary copy of the book to take the playbook with you.
13:30 - 14:30
Lunch Break
14:35 - 15:20
Panel Discussion: Designing Connected Devices with Security Built In
Security is critical for connected embedded devices, and this panel takes a deep dive into how to build it in from the very beginning. Experts discuss best practices including hardware root of trust, secure boot processes, strong encryption, and continuous threat monitoring to protect IoT systems at every stage. The conversation also covers how to stay ahead of evolving cyber threats and the importance of designing security measures that adapt as devices age. Attendees will gain practical insights into creating resilient, long-lasting IoT deployments that can withstand real-world challenges.
15:25 - 15:45
Presentation: From Cloud AI to Physical AI: Why Intelligence Must Move to the Edge
AI is undergoing a fundamental paradigm shift from centralized cloud computing to distributed intelligence at the edge and ultimately to Physical AI, where intelligence is embedded directly within the physical world.
This transformation is driven by the need for real-time perception, low-latency decision-making, and continuous interaction with dynamic environments. As a result, AI is no longer confined to data centers but is becoming an integral part of vehicles, robotics systems, and intelligent infrastructure.
Major technology leaders such as NVIDIA, Tesla, and Apple are converging on this direction by building end-to-end stacks that integrate advanced hardware, edge computing, and on-device AI acceleration.
Ultimately, this shift represents a new phase of computing—where AI is no longer abstract or remote, but physically present, context-aware, and capable of acting within the real world.
