no. the backpack is my sheath. its sword, my macbook, has out king-made excalibur and magicked more than the elder wand. it’s is a moma-worthy bauhaus piece milled by octopi robots from a block of metal. i love how it catches the sun. everyday we portal into central park-sized supercomputers, we tilt orbiting spacecraft to locate food, we peruse 1000000 alexandrian libraries… i bring my backpack to the park without thought of work, would i leave my hands at home?
@amasad@DesarajuSasank In firing patterns, yes. but morphologically each neuron encodes thousands of analog weights. Dendritic compartments, nonlinear integration zones, channel densities, plasticity rules, time-varying states. Could easily push effective parameters per neuron into thousands, not tens.
🚨 New: We built @a16z's personal GPU AI Workstation Founders Edition
- 4x NVIDIA RTX 6000 PRO Blackwell Max-Q (384GB total VRAM)
- 8TB of NVMe PCIe 5.0 storage
- AMD Threadripper PRO 7975WX (32 cores, 64 threads)
- 256GB ECC DDR5 RAM
- 1650Watts at peak (runs on a standard 15Amp/120V circuit).
For training, AI research, and deploying models locally. A datacenter-class AI rig you can keep under your desk.
We are planning to make a limited number of these a16z AI Workstations.
Build guide + how you can make your own 👇
We are not the first to face challenges that are met with the term ‘impossible.’
For perspective, in 1958, the CORONA project returned satellite images from orbit by dropping film canisters to be caught midair by aircraft. It worked for over a decade.
In comparison, SpaceComputer’s challenges are easier, and our timeline to feasibility is shorter. This is a solvable problem; it might just take 𝕦𝕟𝕔𝕠𝕟𝕧𝕖𝕟𝕥𝕚𝕠𝕟𝕒𝕝 𝕤𝕪𝕤𝕥𝕖𝕞𝕤 𝕒𝕟𝕕 𝕤𝕠𝕝𝕦𝕥𝕚𝕠𝕟𝕤.