Скачать или смотреть Shoulda, Woulda, CUDA with Jay Dawani , CEO @ Lemurian Labs | Beyond CUDA Summit 2025

Jay Dawani, CEO of Lemurian Labs, shares his compelling journey from robotics and early work at OpenAI to pioneering a radical new approach in AI software infrastructure. After initially aiming to revolutionize hardware, Jay explains why Lemurian Labs pivoted to building a dynamic runtime and compiler stack, designed specifically to overcome Nvidia’s CUDA monopoly and unlock massive performance gains on platforms like AMD’s MI300.

𝘾𝙤𝙣𝙣𝙚𝙘𝙩 𝙬𝙞𝙩𝙝 𝗝𝗮𝘆 𝗗𝗮𝘄𝗮𝗻𝗶 -
  / jaydawani  

📢 Let us know your thoughts in the comments

--
Timestamps

0:00 - 0:35 - Introducing Jay Dawani, CEO of Lemurian Labs
0:36 - 1:16 - Jay’s Journey: Robotics, OpenAI, and Beyond
1:17 - 2:21 - Why Lemurian Labs Pivoted from Chips to Software
2:22 - 3:15 - Accelerated Software: Unlocking the Next Compute Paradigm
3:16 - 4:37 - Overcoming CUDA: Building a New Compiler & Runtime Stack
4:38 - 5:07 - Performance Gains on AMD MI300 and Nvidia GPUs
5:08 - 6:24 - Motivation Behind Breaking Nvidia’s CUDA Monopoly
6:25 - 7:03 - Jay’s Vision: Democratizing AI for Economic Growth

---
About TensorWave:

TensorWave is the AI and HPC cloud purpose-built for performance. Powered exclusively by AMD Instinct™ Series GPUs, we deliver high-bandwidth, memory-optimized infrastructure that scales with your most demanding models—training or inference.

--
Connect with TensorWave:

https://www.tensorwave.com
https://www.x.com/tensorwavecloud
  / tensorwave  
  / tensorwave_cloud  
   / @tensorwavecloud  



#AICompute #GPUs #BeyondCUDA #AIInfrastructure