I'm a systems engineer in Jaipur.
I build inference-engine at . Go and
vLLM serving models on GPUs, OpenTelemetry into
ClickHouse, Prometheus and Grafana on top.
I also maintain ultrabalancer, an HTTP/2 load balancer in Rust that holds 500k requests a second.
Mostly Rust, Go and TypeScript, plus C and Zig when a problem gets low enough.
Before this: routing.run, an LLM gateway I co-founded and ran through its acquisition, and backend platforms at MegaLLM.