Lightmatter CEO explains how spare lasers and optical switching route around failures in AI systems
BenBajarin · x · 2026-10-02
Ben Bajarin of The Diligence Stack interviews Lightmatter founder and CEO Nick Harris on optical interconnect for AI systems and the manufacturing bottlenecks that will set the pace of adoption.
Key points:
- Demand isn't the constraint—everyone would adopt optics today. The hard question is whether the supply chain can ship qualified optical engines at repeatable yield, as copper hits physical limits at higher link speeds and larger scale-up domains.
- Laser manufacturing: most commercial lasers come from a small group of suppliers using small indium phosphide wafers. Harris argues Lightmatter can move more of that work onto 300mm silicon in mature CMOS foundries, opening a different path to laser capacity.
- Reliability: Lightmatter's Guide uses spare lasers and optical switching to route around a failed laser, addressing the reliability challenge of thousands of optical links.
The full interview is behind a paywall.
Related event: Lightmatter CEO Details Optical Interconnect Roadmap(3 posts)→
More from Infra
- a16z: every $100 into AI buildout sends $50 to chips, $20 to power; 100+ charts — demian_ai · 2026-10-02
- Workload-aware inference: why batch LLM pipelines should plan queries like databases do — sh_reya · 2026-10-02
- Where does GPU spend actually go: training, inference, or idle reserved capacity? — Borges_Engineer · 2026-10-02
- HBM4 controllers eat ~16% of Nvidia's Rubin die, sparking optical interconnect debate — BenBajarin · 2026-10-02
- Modal Clusters goes GA: instant RDMA-connected GPU nodes billed by the second via one decorator — josh_wills · 2026-10-02
- Benchmark Backs Chip Startup Tendrils Compute in Round That Could Top $1B Valuation — Sethwinterroth · 2026-10-02