Developer Workshops at the AI Hardware & Edge AI Summit
Join us on the afternoon of day 2 (Wednesday, Sept 11) for a series of interactive, hands-on learning opportunities, from some of the industry elite.
These workshops are a great way to receive tutorial and demos in a variety of disciplines.
Interested? Purchase a full ticket today and on registration and select which workshops you are most interested in on the registration form. We will then reach out to you to if you have been selected for your chosen workshop.
Alternatively, purchase an expo ticket and upgrade to add workshops for guaranteed access.
Workshop Agenda
SambNova Systems will be hosting 1 workshop (running for 1 hour).
Level: Intermediate
[1:15pm] Workshop = Get Started Developing on the Fastest AI Platform
In this workshop, you will learn how to build LLM-based apps, such as a question-answering system with RAG, in LangChain using Llama-3 at 1,000 tokens per second on the SambaNova AI Platform.
SambaNova delivers generative AI capabilities to the enterprise. In this workshop, you will learn:
- About SambaNova’s full-stack generative AI platform, powered by the SN40L AI chip and delivering unparalleled performance for training and inference
- Samba-1, a trillion parameter composition of experts (CoE) model, and how it can be used for enterprise settings
- How to build and deploy a question-answering app end-to-end with retrieval augmented generation (RAG) for enterprise search using the following suite: LangChain as a framework, Unstructured for pre-processing text documents, E5-large-v2 embedding, ChromaDB vector store, and Llama-3-8B-Instruct running at speed record of 1,000 tokens per second via SambaNova.
AMD will be hosting 2 workshops (running for 1 hour).
[1:15pm] Workshop 1 = "Developing Generative AI Applications for AI-Powered PCs"
- This hands-on workshop guides participants in developing agentic RAG applications using GAIA, a generative AI sandbox for AI-powered PCs. Attendees will explore building on LLM frameworks like llama-index and learn the advantages of running large language models locally on personal computers.
[2.45pm] Workshop 2 = "AMD x Hugging Face : Deploying optimized LLMs on AMD GPUs"
Join this interactive workshop and learn how AMD and Hugging Face collaborated to optimize and deploy LLMs using ROCm software and the benefit of end-to-end workflows in an open-source development ecosystem.
Mobilint will be hosting one workshop (running for 1 hour).
[2:30pm] Workshop = "Exploring AI Chips, Boards, and SDKs for Successful On-Device and On-Prem AI"
In this workshop, we will showcase two AI chips, ARIES for on-premises AI and REGULUS for on-device AI, with the following features:
Versatility: Supports over 200 models.
- Power Efficiency: Live demo of MLA100 running 30+ deep learning models on a single NPU with only PCIe power and a small cooler.
- PTQ Accuracy: Over 99% FP32 accuracy without retraining, while maintaining INT8 efficiency.
- Scalability: Supports 1 TOPS to 100+ TOPS with a scalable NPU architecture and software stack.
- SDK Ease of Use: User-friendly SDK enables easy integration and deployment of pre-trained models with minimal code.
Qualcomm will be hosting 1 workshop (running for 1 hour).
[3:45pm] Workshop = Accelerating On-Device AI Development with Qualcomm AI Hub
- Want to easily deploy ML to edge devices? In this workshop, address the common challenges faced by developers migrating AI workloads from the cloud to edge devices. Qualcomm aims to democratize AI at the edge, easing the transition to the edge by supporting familiar frameworks and data types.
- Developers can follow along, gaining knowledge and tools to efficiently deploy optimized models on real devices using Qualcomm AI Hub. Qualcomm will walk through how to get started using Qualcomm AI Hub, iterate on your model and meet performance requirements to deploy on device! They'll show examples on how to optimize models and bundle the downloadable target asset into your application
Nscale will be hosting 1 workshop (running for 1 hour).
- Agenda to include: Understanding GPU Technologies, Performance Analysis, and an interactive Q&A session providing expert insights and advice to help you select the optimal GPU for your needs...
Arete Research will be hosting 1 investor workshop (running for 45 minutes).
[4:45pm] Workshop = AI Hardware Market Analysis: Analysing The Golden Decade Ahead
- Led by Brett Simpson, Arete Research will layout a case for how the AI hardware market TAM evolves in the 2020s and why the semis roadmap for AI (leading edge manufacturing, compute, memory, interconnect, switching) needs to rearchitect for low-power. We also explain why hyperscaler collaboration is long overdue and the challenges ahead taking on established AI players (such as Nvidia) in the years to come.
INTEL MINI SUMMIT WORKSHOP TRACK
Intel will be hosting 3 workshops (running for 1 hour).
[2:30pm] Workshop 1 = "Getting Started on Intel® Gaudi®"
- In this session, Intel will provide a hands-on introduction to Intel® Gaudi®, where you'll be able to start running Generative AI models from Hugging Face for both Inference and Fine Tuning. Intel will show you the hardware and software details. Participants will also be exposed to the additional tutorials and source code.
[3:30pm] Workshop 2 = "How to use Retrieval Augmented Generation (RAG) on Intel® Gaudi®"
- Intel will provide hands-on details on how to setup and execute a Retrieval Augmented Generation (RAG) workflow taking advantage of the power of the Intel® Gaudi® accelerator. The Intel team will show you how to use the Hugging Face Text Generation Inference (TGI-Gaudi) and Text Embedding Inference to set up the RAG components and run with a Large Language model.
[4:30pm] Workshop 3 = "Deploying Personal AI Assistants with Voice, LLMs, RAG via OpenVINO™ on the AI PC "
- Explore how to build and deploy local LLM-based assistants on the AI PC or at the edge. Our pipeline leverages a real-time speech transcription model (Distil-Whisper), and Large Language Model (LLama3) powered chatbots leveraging Retrieval Augmented Generation (RAG), to personalize user interactions via text generation and summarization over prior interaction history. We discuss how the Intel® Core™ Ultra enables efficient deployment of LLMs on the CPU, iGPU, and NPU, through optimization techniques such as quantization and OpenVINO™ compression libraries. Live demos will be presented throughout the session to ensure developers can see the work in action.
Workshop Speakers
Kalin Ovtcharov
Kalin leads GenAI solutions including Agentic RAG, and LLM based reference solutions to support and inform customer engagements at AMD. Previously he was the CEO & Co-Founder of Extropolis where he led a team developing cloud/mobile products for Agentic RAG, ASR, Stable Diffusion, and ChatBot applications. Prior to Extropolis, Kalin was a researcher and principal lead at Microsoft where he developed and launched configurable cloud-scale AI Inference solutions with Project Brainwave.
Seungrok Jung
Seungrok Jung is an AI/ML engineer who brings more than 15 years of experience in AI software engineering and AI accelerator hardware design. At AMD, he specializes in optimizing and benchmarking performances of machine learning algorithms on AMD devices. His interest lies in analyzing recent ML models and extracting best performance of the models on different choices of ML frameworks, libraries, GPU kernels and toolkits. He is also interested in contributing to ML open-source communities so that everyone can enjoy the full potential of the AMD devices.
Before joining AMD, Seungrok worked on AI chip accelerator and Mobile SoC processor design at the corporate R&D center of SK Telecom and Samsung Electronics. He holds a master’s degree in electrical engineering from the Korea Advanced Institute of Science and Technology.
Greg Serochi
Principal AI Technical Program Manager and Ecosystem Enabling Lead for Intel Gaudi. As the Ecosystem Enabling Lead, spearheaded the strategy and execution of Intel Gaudi's ecosystem collateral, customer training and positioning. Greg's role is to create content and training that make it easier for customer to use Intel Gaudi.
Raymond Lo
Elio Van Puyvelde
Elio has over 15 years of entrepreneurial experience in HPC and AI industries; Co-Founder of ElioVP, a manufacturer, designer, and operator of premium HPC hardware and software systems.
Dongjoo Shin
Kian Mohadjerin
With an extensive background in ML, Python, and HIP/C++, Kian Mohadjerin leads AI innovation at Nscale. He has developed a deep expertise through hands-on development and leadership in various AI initiatives.
Jörg Roskowetz
Jörg Roskowetz is an experienced professional in the field of AI and Web3 technologies. He collaborates closely with key strategic partners and his role involves spearheading the development of innovative proof of concepts and bespoke graphic features tailored to meet the evolving demands of the industry.
Vasanth Mohan
Vasanth Mohan is a seasoned technology leader, serving as Director of Technical Product Marketing and Developer Relations at SambaNova. With a proven track record of empowering thousands of developers and startups across emerging technologies, including VR/AR, Edge Computing, and AI, Vasanth has helped numerous solutions successfully deploy in the enterprise. At SambaNova, he is now focused on enabling developers and the enterprise to unlock the value of Agentic AI to solve today's business challenges.
Varun Badrinath Krishna
Varun is a Senior Principal AI Solutions Engineer at SambaNova Systems. He is currently investigating the benefits of fine-tuning embedding & decoder LLMs in retrieval augmented generation (RAG). Previously, he led the deployment of AI/ML applications across CRM, e-commerce, healthcare, finance, energy, manufacturing, fraud detection, and cyber security at Fortune 500 enterprises. He has worked at C3.ai, Cisco, IBM Research, ABB, and research institutes in Singapore. He holds a Ph.D. in Computer Engineering from the University of Illinois at Urbana-Champaign.
Brett Simpson
Brett is a co-founder of Arete (formed in 2000) and is based in the firm's London office. He focuses on the global semiconductor component sector. Brett is a regular public speaker at industry events and after 17 years looking at the sector, has a wealth of experience to draw on. Prior to Arete, Brett spent two years at Goldman Sachs in an equity analyst role, specialising in European technology following three years with Ericsson UK, working in business development, covering all aspects of wireline and wireless telecom infrastructure.
Joel Hagberg
Joel leads the Intel® RealSense™ Product Management, Marketing and Customer Support Teams. Joel leads product management for development of Computer Vision HW/SW with a focus on AI Applications and AI inference at the edge. Joel joined Intel RealSense in 2018 and has over 25 years of experience leading Product Management, Product Marketing and Business Development teams. Prior to Intel, Joel was VP of Products at Seagate Technology. He was VP of Marketing for Toshiba America’s Storage Division, and Fujitsu’s VP of Marketing & BusDev in Storage, Robotics, Logistics, Biometrics, Healthcare and Digital Signage.
Gustav Larrson
Gustav did his Ph.D. in Computer Science at the University of Chicago, focusing on Computer Vision and Deep Learning. He has spent half of his career developing and shipping ML applications (e.g., camera algorithms used in trillions of photos every year), experiencing first-hand the pain points of on-device deployment. The other half has been spent developing ML infrastructure to mitigate those pain points and empower developers.