Thomas Sohmers

Mentioned in 1 episode

"Semi Analysis has their Agent X benchmark based on taking a whole lot of Claude Code sessions with dozens to hundreds of turns and sub-agents. What they found is about 96% of all tokens that go through these entire sessions are cached."

Mentioned in 1 episode

"With our upcoming generation, we're going to have eight times more memory capacity than the highest memory SKU from Nvidia. And Nvidia is actually decreasing the amount of memory per device based on the market memory conditions."

Mentioned in 1 episode

"Deep Seek beginning of 2025 with Deepseek V3 made a lot of waves because they were able to get a massive decrease in KV cache size with multi-head latent attention."

Mentioned in 1 episode

"Anthropic and OpenAI have better uses of their capital and mental power than doing all of that Scale AI work."

Mentioned in 1 episode

"The smarter and more capable that Siri is on my phone is going to result in it doing a whole bunch of background tasks and things that remove me from having to be the one that instigates it having requests and data be processed by even smarter models."

Mentioned in 1 episode

"The computer usability with a lot of generic tools — being able to do Blender animations — there's a bunch of memes online of it recreating different videos. The fidelity of that was basically impossible with GPT-5.6."

Mentioned in 1 episode

"Everything from ChatGPT, Claude, etc. — how does the infrastructure required for inference what you're working on change compared to training?"

Mentioned in 1 episode

"A company we're partnered with and I'm good friends with the CEO is called Panthalassa, building ocean-based data centers — basically a very interesting pumped hydro solution in the middle of the ocean. There are alternatives that don't require going to space."