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

Improve Instagram comment ranking

Instagram comments are currently ranked chronologically. Design an ML ranking system that improves comment quality and relevance while handling safety, freshness, personalization, and feedback loops.

hardsystem designEvidence 77/1001 source reportMeta

The 60-second answer

Frame the goal as useful, relevant, safe comments—not raw engagement. Generate eligible candidates, then rank with user/comment/author/post interactions, content semantics, freshness, quality, safety, and conversation context.

Build the answer in this order

1
Frame the problem

Frame the goal as useful, relevant, safe comments—not raw engagement.

2
Design the data path

Generate eligible candidates, then rank with user/comment/author/post interactions, content semantics, freshness, quality, safety, and conversation context.

3
Choose the modeling stack

Correct for position-biased implicit labels and evaluate ranking quality together with hides/reports, diversity, latency, and creator/user outcomes.

4
Serve, evaluate, iterate

Run guarded experiments and monitor feedback loops so already-prominent comments do not monopolize future training labels.

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

  • Discuss multi-objective ranking, exploration, position-bias correction, and causal evaluation.
  • Treat integrity/safety constraints as first-class ranking requirements, not an afterthought.

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.