Design and explain robust web APIs for ML inference
NVIDIAData Scientist · HR Screen
Last updated
This question evaluates a candidate's proficiency in designing HTTP REST APIs for ML inference, including API endpoints, versioning, idempotency, request/response schemas, authentication/authorization, rate limiting, observability, and backward-compatibility policies, and it falls under the Coding & Algorithms domain for a data scientist role.
Design an HTTP API to serve image-based model predictions. Include: 1) Endpoints (e.g., POST /v1/predict, GET /v1/jobs/{id}/status), methods, idempotency, and versioning. 2) Request/response schemas (JSON + multipart), content types, standard error codes, and retry semantics with exponential backoff and idempotency keys. 3) Authentication/authorization (OAuth2/OIDC with scopes), rate limiting/quotas, and audit logging. 4) Backward compatibility and a deprecation policy. 5) Security (TLS, input validation, JWT verification), PII handling, and observability (structured logs, metrics, tracing, request IDs). 6) Provide a concise OpenAPI 3.0 snippet for one endpoint that captures parameters, schema, and error responses.
Solution walkthrough is for members
Get the full walkthrough with reference code and complexity analysis, plus Kitty Coach at 300 messages a day. Hints, the code runner and community answers stay free.