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

Design a short-video recommendation system similar to Reels.

Design a short-video recommendation system similar to Reels.

hardsystem designEvidence 70/1001 source reportMeta

The 60-second answer

Clarify product objective, prediction/decision target, latency and scale, freshness requirements, and the relative cost of false positives, false negatives, and abstention. Define data sources, labels, leakage controls, feature pipeline, baseline, and model family before proposing a complex architecture.

Build the answer in this order

1
Frame the problem

Clarify product objective, prediction/decision target, latency and scale, freshness requirements, and the relative cost of false positives, false negatives, and abstention.

2
Design the data path

Define data sources, labels, leakage controls, feature pipeline, baseline, and model family before proposing a complex architecture.

3
Choose the modeling stack

Choose offline metrics tied to the product decision, then specify online experimentation, monitoring, drift detection, and rollback criteria.

4
Serve, evaluate, iterate

Cover serving, cold start, feedback loops, fallbacks or human review where relevant, and how production outcomes feed retraining.

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

  • Explicitly address delayed/biased labels and cases where the model changes the data it later learns from.
  • Version data, features, model, and policy together so incidents are reproducible and safely reversible.

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.