Sunay Hegde — Co-Founder

I built Air.rs to run 70B models — which need ~35GB VRAM at Q4 — on an RTX 4090 with only 24GB, using a triple-buffered pipeline where the GPU executes layer N while PCIe uploads layer N+1. I gravitate toward low-level systems, algorithms, and anything at the intersection of performance and intelligence. The language matters less than the problem — though lately that language is Rust. I'm always open to: — Open source collaborations in AI infrastructure and systems — Conversations with researchers and engineers building real models — Opportunities where technical depth actually matters

Stackforce AI infers this person is a specialist in AI infrastructure with a focus on performance optimization.

Location: Mumbai, Maharashtra, India

Experience: 0 mo