Super Data Science: ML & AI Podcast with Jon Krohn 881: Beyond GPUs: The Power of Custom AI Accelerators, with Emily Webber
62 snips
Apr 22, 2025 
Emily Webber, Principal Solutions Architect at AWS, shares insights into cutting-edge AI hardware. She discusses the impressive specs of Trainium and Trainium2, highlighting how they outperform traditional GPUs. Learn about the AWS Neuron SDK and its customization options for AI frameworks. Emily explains the Nitro System's role in enhancing cloud performance and security. Listeners will discover the criteria for choosing between AWS’s accelerators and the importance of tailored AI infrastructure for effective model training.
AI Snips
Chapters
Books
Transcript
Episode notes
Meditation Boosts Coding Focus
- Emily found Buddhist meditation and focus practice greatly enhanced her problem-solving in computer science.
- She emphasizes mental calmness as a key tool for coding and complex algorithm development.
Annapurna Labs' Cloud Revolution
- AWS values customer choice in hardware like CPUs and AI accelerators to optimize cloud performance.
- Annapurna Labs revolutionized cloud computing with Nitro system and custom chips like Graviton and Trainium.
Trainium 2's Massive Upgrade
- Trainium 2 offers 4x more compute by quadrupling neuron cores and memory per card.
- Ultra server combines multiple Trainium 2 instances enabling massive scalable training for large language models.
