Code Ahead of the Curve with FREE Technical Trainings

Large language models and the applications built around them have emerged as powerful tools for understanding and generating natural language. However, optimizing for these models for maximum efficiency and performance remains a significant challenge.

This session introduces a solution: How to optimize LLM workloads on target hardware using the Intel extensions for transformers and PyTorch.

Key takeaways:

1. Introduction to Intel® Extension for Transformers and Intel® Extension for PyTorch, two powerful libraries for enhancing AI workload performance on Intel platforms.
2. Using the PyTorch extension’s API calls to optimize LLM performance and memory usage.
3. Using the transformers extension’s optimization features such as model compression, neural speed, and neural chat, which is a framework to build customized chatbots.

Sign up today.

Skill level: Novice

Featured software

Download the following, per your preference:

• Intel Extension for PyTorch from GitHub or from the Intel AI Tools selector
• Intel Extension for Transformers (GitHub)

Download code samples

• Getting started with Intel Extension for PyTorch
• See all code samples

Kevin Ta
AI Software Solutions Engineer, Intel

Pramod Pai
AI Software Solutions Engineer, Intel

If you’ve ever considered partnering with Intel or wondered what that might be like, this session is for you.

First, you’ll get a detailed overview of the Intel-Bilic alliance, wherein the AI-based cybersecurity startup significantly transformed its business by leveraging Intel expertise, entrepreneurial opportunities, and HW/SW technologies.

Second, Bilic will showcase their innovative Bilic Verify solution, built using the advanced AI software and hardware available in Intel® Tiber™ Developer Cloud¹.

Topics include:

• A breakdown of Bilic’s journey with Intel and how it has reshaped their LLM-driven solutions and services, boosting LLM performance and accuracy, cost efficiencies, and customer experience.
• How its use of Intel Tiber Developer Cloud—AI models, hardware, and compute power—has significantly improved customer approvals and onboarding processes for Bilic Verify’s core offering: helping companies run reliable, easy, secure, and instant background checks to make informed hiring decisions.
• A live demo of their advanced AI-based solution, including best practices for utilizing the full features, capabilities, and processing power of Tiber Developer Cloud.

Sign up today.

Skill level: All

Featured software platforms

• Bilic Verify: Learn more | Sign up
• Intel Tiber Developer Cloud: Learn more | Sign up

¹Formerly Intel® Developer Cloud

Saminu Salisu
Blockchain Developer, Data Scientist, Co-founder, Bilic, Ltd.

What is an AI PC and how do developers exploit its AI-acceleration capabilities across the included CPU, GPU, and NPU?

This session delivers answers and unpacks fundamental and advanced techniques for tapping into the ever-expanding potential of AI using OpenVINO on the Intel AI PC

Key takeaways:

• An overview of Intel’s AI PC
• Approaches for making current-state and next-generation AI and GenAI models more performant and power-efficient
• Why you should consider running AI and GenAI models on client and edge devices, including tips on low-power implementations.
• Techniques for optimizing and deploying AI applications on the AI PC’s different compute engines using OpenVINO.
• How to access and use the OpenVINO Notebooks repository, including its functionalities and applications.

Includes live demos showcasing how to seamlessly transition AI/GenAI apps across compute engines with OpenVINO for popular use cases, such as background blurring on video calls, object detection, and GenAI-powered image generation and chatbots.

Sign up today.

Skill level: All

Featured software

• Download the OpenVINO toolkit
• Explore Hugging Face Optimum-Intel Inference with OpenVINO

Intel, the Intel logo, OpenVINO, and the OpenVINO logo are trademarks of Intel Corporation or its subsidiaries.

Ria Cheruvu
Blockchain Developer, Data Scientist, Co-founder, Bilic, Ltd.

Model size + limited hardware resources in client devices (i.e., disk, RAM, CPU) make it increasingly challenging to deploy LLMs on laptops compared to cloud-based solutions.

The Intel AI PC solves this issue by including a CPU, GPU, and NPU on one device.

This session focuses on the NPU, showcasing how to prototype and deploy LLM applications on it locally.

Key learnings:

• How NPU architecture works, including features, advantages, and capabilities in accelerating neural network computations on Intel® Core™ Ultra processors (the backbone of Intel’s AI PCs).
• Practical aspects of deploying performant LLM apps on Intel NPU—from initial setup to optimization and system partitioning—using OpenVINO toolkit and its NPU plugin.
• Large language modes: what they are and advantages/challenges of local inference.
• Fast LLM prototyping on Intel Core Ultra processors using the Intel NPU Acceleration Library.

Includes real-world examples and case studies (like chatbots and RAG) that showcase the seamless integration of LLM applications with NPUs, including how this synergy can unlock performance and efficiency.

Sign up today.

Skill level: All

Featured software

• Download the OpenVINO toolkit
• Explore Hugging Face Optimum-Intel Inference with OpenVINO
• Intel NPU Acceleration Library

Intel, the Intel logo, OpenVINO, and the OpenVINO logo are trademarks of Intel Corporation or its subsidiaries.
Intel, the Intel logo, and Intel Core are trademarks of Intel Corporation or its subsidiaries.

Alessandro Palla
Machine Learning Engineer, Intel Corporation

Large language models promise to revolutionize how enterprises operate, but making them production-ready means solving for privacy risks, security vulnerabilities, and performance bottlenecks.

Not so easy.

This session focuses on how AI startup Prediction Guard found a solution to these challenges by using the processing power of Intel® Gaudi® 2 AI accelerators in the Intel® Tiber™ Developer Cloud¹.

Topics include:

• Prediction Guard’s pioneering work hosting open source LLMs like Llama 2 and Neural-Chat in a secure, privacy-preserving environment with filters for PII, prompt-injection attacks, toxic outputs, and factual inconsistencies.
• How they optimized batching, model replication, tensor shaping, and hyperparameters for 2x throughput gains and industry-leading time-to-first-token for streaming.
• Architectural insights and best practices for capitalizing on LLMs.

Sign up today.

Skill level: Expert

Featured software

This session showcases the Intel Tiber Developer Cloud: Learn more | Sign up

Download code samples

• Intel-Hugging Face Neural-Chat 7B
• See all code samples

1 Formerly Intel® Developer Cloud
Intel, the Intel logo and Gaudi are trademarks of Intel Corporation or its subsidiaries.

Daniel Whitenack
Prediction Guard