LLM gateway
Multi-model routing, caching, and call logs.
Go · Redis · LLM APIs
multica.hongbos.com
Usage observability
Tokens, latency, spend — usage made visible.
TypeScript · Postgres · Charts
usage.hongbos.com
Cluster watch
K8s and uptime checks for the home cluster.
K8s · Prometheus · Docker
kuma.hongbos.com
RAG & evals, applied on live systems.
Go LLM gateway — routing & caching.
The layer where models meet infra.