Sliding Windowhard70 minAI/ML Engineer

Ranking Streaming Window Anomaly

AI/ML Engineer signal: sliding window + frequency map in a learning-to-rank pipeline context. This is a ProdMatch-owned ai ml engineer drill, framed as a May 2026 Dream11 Search ML simulation, not a copied platform question.

Company context

Dream11 · Search ML

Freshness

May 2026

Product surface

learning-to-rank pipeline

ProdMatch interview simulation based on product-team patterns; not a claim of a real company question.

Question

For learning-to-rank pipeline, process a timestamp-ordered stream of documents. Find the longest contiguous window where at most k distinct risk labels appear and the total severity stays under budget.

Input

Array of events {label, severity} and integers k, budget.

Output

Maximum window length satisfying both constraints.

Constraints

1 <= events.length <= 300000
1 <= k <= 50
Severity values are positive integers.

Concepts

vector search
RAG retrieval
recommendation graphs
sliding window
frequency map
stream processing

labels: A,B,A,C with k=2 and budget=7 -> longest valid window length is 3

Approach

Try framing your own approach first. The 30 seconds you think before peeking is where learning happens.

Clean Solution

Reveal the approach first.

How well did you understand?

Your rating tunes when this problem shows up again.

Common Mistakes

Update the answer only after the window is valid.
Avoid recomputing the whole window on every move.

Next Similar Problems

Vector Streaming Window Anomalyhard GraphRAG Streaming Window Anomalyhard Recommend Streaming Window Anomalyhard