Domain
Semiconductors
7 episodes
-
Neoclouds become AI’s new power brokers
Onyx and Echo dig into the rise of neoclouds after the Anthropic-Volta deal, treating it as a signal that AI infrastructure is splitting into specialized capacity markets. They debate who these providers actually serve, why scarcity is the real business model, and where enterprises risk repeating the same rushed-cloud mistakes in a more expensive form.
-
Compute Forecast — AI 2027
Vince and Ava dig into Romeo Dean’s 2025 “Compute Forecast — AI 2027” and tease apart which parts of the compute story feel grounded (10x global AI-relevant compute, concentration in a few labs) versus which jumps (a million “superintelligent” research agents at 50x human speed, three and a half percent of U.S. power) feel more like scenario fiction. They map the technical assumptions behind H100-equivalent growth, utilization, and chip efficiency to actual product and research decisions, and argue that the real takeaway isn’t “AGI by 2027” but “whoever owns the scheduler and the power bill sets the rules.”
-
What is IBM’s nanostack chip architecture?
IBM announced a new sub-1 nanometer nanostack chip architecture that stacks transistors vertically instead of horizontally, promising nearly double the transistor density of current 2nm chips. Cody leads skeptically: the announcement is a capability claim without shipping proof, and the fabrication challenges—wafer-to-wafer bonding precision, High NA EUV maturity, and unknown yield at volume—are enormous. Justy pushes back: the material-decoupling unlock (optimizing n-type and p-type transistors independently) is real, and for AI accelerators, the power efficiency gains directly address data-center bottlenecks. They land on a shared reading: IBM's architecture is mechanistically sound and the roadmap credible, but this is a research milestone, not a product—and the gap between lab demo and foundry-scale manufacturing is where most announcements die.
-
OpenAI and Broadcom unveil LLM Optimized inference chip
OpenAI and Broadcom unveil Jalapeño, a custom AI inference chip designed for LLM workloads, promising substantial performance-per-watt improvements.
-
AMD Delivers Breakthrough MLPerf Training 6.0 Results
AMD's MLPerf Training 6.0 results show significant performance gains, including a 3.5X generational leap on Llama 2-70B and competitive performance on core LLM workloads, with a focus on multi-node training and platform readiness.
-
Netflix Uncovers Kernel Level Bottlenecks While Scaling Containers on Modern CPUs
Netflix discovered that scaling hundreds of containers simultaneously hits deep kernel-level bottlenecks in the Linux virtual filesystem, where thousands of mount operations create lock contention that varies dramatically across different CPU architectures. Their solution involved redesigning overlay filesystems to reduce mount operations from O(n) to O(1) per container.
-
China unveils world's cheapest humanoid robot under $1,400
The unveiling of Noetix's Bumi, the world’s cheapest humanoid robot at $1,370, is a game-changer in robotics and education. Hosts delve into its features, potential uses, and the broader implications for society.