CrackML by @ml.with.umang
Interview questions / ML System Design
ML System Design interview question

Design Global Search for Ephemeral Status Posts

Design search over short-lived status-style posts where freshness and deletion semantics matter.

hardsystem designEvidence 41/1001 source reportMeta

The 60-second answer

Define query types, visibility/privacy, freshness SLA, expiration/deletion semantics, and ranking objective before choosing retrieval technology. Maintain a rapidly updated lexical/vector index with strict ACL filtering and expiration propagation; combine semantic relevance with recency and social/context signals.

Build the answer in this order

1
Frame the problem

Define query types, visibility/privacy, freshness SLA, expiration/deletion semantics, and ranking objective before choosing retrieval technology.

2
Design the data path

Maintain a rapidly updated lexical/vector index with strict ACL filtering and expiration propagation; combine semantic relevance with recency and social/context signals.

3
Choose the modeling stack

Use candidate retrieval followed by ranking, with caches that respect visibility and deletion invalidation.

4
Serve, evaluate, iterate

Evaluate recall/relevance, freshness lag, deletion correctness, latency, and abuse/safety slices.

A useful interview mental model

This is the shape of a strong answer—not a script to memorize.

01Requirements
02Data
03Model / Retrieval
04Serving
05Monitor

Senior-level signal

  • Senior answers treat privacy and deletion as correctness requirements, not post-ranking filters.
  • Discuss index backfill, hot-key/fanout behavior, and how to avoid stale cached results after privacy changes.

What the interviewer is really testing

Product framing, data design, modeling choices, serving constraints, reliability, evaluation, and explicit trade-offs.

Likely follow-up questions

What changes at 10× traffic or data volume?
Which failure mode would you monitor first in production?
How would you evaluate this offline and online before rollout?

Common weak-answer patterns

  • Jumping to a model before defining the product contract.
  • Listing components without bottlenecks, metrics, or failure handling.
  • Ignoring data quality, serving latency, monitoring, and iteration.