tezvyn:

Design a multi-tenant model serving platform

Source: interviewintermediate

WHAT IT TESTS: multi-tenant ML serving design. OUTLINE: share infrastructure to cut cost while enforcing tenant data isolation, fair resource allocation against noisy neighbors, and per-tenant performance via quotas and autoscaling.

WHAT IT TESTS: whether you can balance cost-saving sharing against isolation and fairness. ANSWER OUTLINE: pack multiple tenants onto shared serving infrastructure to raise utilization and cut cost, but solve three problems. Security: strictly isolate each tenant's data, models, and requests via authentication, authorization scoped per tenant, and namespace or network separation. Resource isolation: prevent noisy neighbors with quotas, limits, and scheduling so one tenant cannot starve others.

Read the original → interview

Get five bites like this every day.

Tezvyn delivers a daily feed of 60-second tech bites with quizzes to lock in what you learn.

Design a multi-tenant model serving platform · Tezvyn