@briantimar@mihirneal Well, I don't think we'll have a high ratio of gpu clusters working on solving ad spend. Hopefully not, at least.
The concentration of talent at meta and Google working on these problems for the past decade should go down as a huge waste of resources
@briantimar@mihirneal Its easier to convince computers to solve hard problems than people. I imagine the compute allocation will be more efficient than human
Retail investors will be able to participate at the same prices as the big institutions. Expected SPCX price of $135 per share → https://t.co/eKBA0tzXbH
SpaceX has almost finished writing V1.0 of an in-house AI training stack in C that exact-maps to 220k GB300s with 800G NICs, making heavy use of pipeline parallelism and getting as close to bare metal as possible.
The potential speed improvement vs JAX for large training runs is over an order of magnitude.
I am super excited to share that @Starcloud_ has raised a $170M Series A at a $1.1bn valuation to fuel our development of data centers in space 🚀
The round comes after the successful deployment of our first satellite, Starcould-1, a few months ago, which had the first @NVIDIA H100 on board and was the first to train an LLM in space. The funds will be used to develop our third satellite, which aims to be cost-competitive with Earth-based data centers in terms of AI inference cost.
The round was led by @Benchmark and @EQT Ventures, and we are excited to welcome Benchmark GP, @Chetanp Puttagunta, to our board. We are also excited to welcome other new investors, including the world's largest infrastructure fund, @Macquarie Capital, @SevenSevenSix 7️⃣7️⃣6️⃣, Manhattan West, Adjacent, Carya, GSBackers, and Harpoon.
We are very grateful for the continued support of existing investors, including @NFX, @NebularVC, @YCombinator, @FUSE_VC, @Soma_Capital, 3Capital Partners, Wyld VC, Tiny VC, and Taurus Ventures.
Onwards!