AI Hardware

concepts · 6 notes linked

Related: Google · Inference Optimization · Nvidia · Fab2 · Jim Keller · Sam Zeloof · Openai · Terafab

Notes

• Google DeepMind pushes for faster ASIC design, energy efficiency to meet AI surge — Google DeepMind accelerating ASIC cycles for AI inference efficiency
• Jim Keller's startup rebrands Atomic Semi as Fab2, moves to Texas — Startup mass-producing small "fab fab" semiconductor factories
• Nvidia CEO: We bet the farm on AI and no one knew it | TechCrunch — Jensen Huang recounts Nvidia's 2018 AI pivot from rasterization to GPU-accelerated ML
• Nvidia is preparing for the post-GPU AI era as it is reportedly recruits ASIC engineers to fend off competition from Broadcom and Marvell — Nvidia entering ASIC market to compete in shift from GPU training to inference chips
• This AI Uses Light Instead of Electricity and It's Mind-Blowingly Fast — Photonic neural network accelerator processes distributed acoustic sensing in real-time
• tinygrad — Minimal neural network framework and companion ML hardware boxes