Insights

From the blog

Engineering deep-dives, customer stories and product updates from Velerion.

Cutting tail-latency for cold inference
Engineering
Engineering2 min read

Cutting tail-latency for cold inference

How we rethought placement and warm-pool eviction to keep cold starts under 100ms.

By Sofia Krause

#Inference#Latency#Scheduling
Read more
From notebook to 24/7 service in a weekend
Customer story
Customer story2 min read

From notebook to 24/7 service in a weekend

A research team's path from a one-off prototype to a multi-region API serving real users.

By Mateo Rossi

#Customer story#Deployment#Multi-region
Read more
Streaming inference, traced end-to-end
Deep dive
Deep dive2 min read

Streaming inference, traced end-to-end

Tracing token-by-token responses across regions without melting your observability bill.

By Amara Diallo

#Observability#Tracing#Streaming
Read more