From the team
The APIHub Star blog
Notes on AI infrastructure, gateway engineering and building reliable systems at scale.
Engineering
Scaling an AI gateway to 14B requests/month
How we route, cache and fail over across providers without adding tail latency.
Product
Adaptive routing is now generally available
Let the gateway pick the cheapest upstream that meets your latency target.
Security
Designing audit logs you can actually trust
Immutable, tamper-evident logging for every request through the gateway.