25 Aug 2026
Signal Headquarters
Vol. I
No. 237

Etched demonstrated exceptional hardware execution speed, going from silicon to rack-running inference in 40 days versus a competitor's 10 months.

The case

Poolside's Laguna XS2 was developed from the start of pre-training to launch in only 5 weeks.

“You looked at Laguna XS2 that we launched it was 5 weeks from the beginning of pre-training to launch.”
Eiso Kant · 22 Jul 2026

Etched iterated through multiple rack designs before their chips returned from fabrication, using thermal chips to develop cold plates.

“We actually went through multiple iterations of the rack before the chips even came back.”
Gavin Uberti · 30 Jun 2026

Topics

AI HardwareAI InferenceSemiconductors

Signal Headquarters · compiled from attributed public discussion. Last updated 2026-07-22.