Design Playlist System

Design Playlist System — Problem Statement

A playlist system is the curation control plane for streaming products: users create named, ordered collections of catalog references (tracks or videos), share them, collaborate on edits, and hand the ordered list to a playback service that resolves media bytes separately.

Interviewers at Spotify, YouTube, and Apple use this prompt to test ordered-list storage, collaborative concurrency, rights-aware materialization, fan-out on updates, and read-heavy caching — not full video encoding or recommendation training.

Scale anchors

• Spotify: 500M+ playlists; some lists exceed 10,000 tracks; collaborative editing on premium tiers
• YouTube: Public playlists drive discovery; very large lists use continuation tokens
• Apple Music: Library sync merges playlist deltas across devices via iCloud

Core journeys

Owner: create playlist → add/reorder items → set visibility → share link.

Collaborator: join shared playlist → concurrent add/remove → resolve version conflicts.

Listener: open playlist → paginated hydrate with catalog metadata → play via playback API.

Platform: ingest takedown → filter items → invalidate caches → notify subscribers.

Extended interview notes

• Problem Statement & Context note 1: Ordered lists are the product primitive — playback is a separate service. Tie decisions to measurable playlist SLIs (read p95 <80ms, reorder p99 <200ms, collab conflict rate) and call out failure isolation.
• Problem Statement & Context note 2: Filter unlicensed items before returning playlist payloads. Tie decisions to measurable playlist SLIs (read p95 <80ms, reorder p99 <200ms, collab conflict rate) and call out failure isolation.
• Problem Statement & Context note 3: Cache playlist headers separately from item pages. Tie decisions to measurable playlist SLIs (read p95 <80ms, reorder p99 <200ms, collab conflict rate) and call out failure isolation.
• Problem Statement & Context note 4: Filter unlicensed items before returning playlist payloads. Quantify assumptions before naming databases.
• Problem Statement & Context note 5: Cache playlist headers separately from item pages. Quantify assumptions before naming databases.
• Problem Statement & Context note 6: Ordered lists are the product primitive — playback is a separate service. Quantify assumptions before naming databases.
• Problem Statement & Context note 7: Filter unlicensed items before returning playlist payloads. Quantify assumptions before naming databases.
• Problem Statement & Context note 8: Cache playlist headers separately from item pages. Quantify assumptions before naming databases.
• Problem Statement & Context note 9: Ordered lists are the product primitive — playback is a separate service. Quantify assumptions before naming databases.
• Problem Statement & Context note 10: Filter unlicensed items before returning playlist payloads. Quantify assumptions before naming databases.
• Problem Statement & Context note 11: Cache playlist headers separately from item pages. Quantify assumptions before naming databases.
• Problem Statement & Context note 12: Ordered lists are the product primitive — playback is a separate service. Quantify assumptions before naming databases.
• Problem Statement & Context note 13: Filter unlicensed items before returning playlist payloads. Quantify assumptions before naming databases.
• Problem Statement & Context note 14: Cache playlist headers separately from item pages. Quantify assumptions before naming databases.
• Problem Statement & Context note 15: Ordered lists are the product primitive — playback is a separate service. Quantify assumptions before naming databases.
• Problem Statement & Context note 16: Filter unlicensed items before returning playlist payloads. Quantify assumptions before naming databases.
• Problem Statement & Context note 17: Cache playlist headers separately from item pages. Quantify assumptions before naming databases.
• Problem Statement & Context note 18: Ordered lists are the product primitive — playback is a separate service. Quantify assumptions before naming databases.
• Problem Statement & Context note 19: Filter unlicensed items before returning playlist payloads. Quantify assumptions before naming databases.
• Problem Statement & Context note 20: Cache playlist headers separately from item pages. Quantify assumptions before naming databases.
• Problem Statement & Context note 21: Ordered lists are the product primitive — playback is a separate service. Quantify assumptions before naming databases.
• Problem Statement & Context note 22: Filter unlicensed items before returning playlist payloads. Quantify assumptions before naming databases.
• Problem Statement & Context note 23: Cache playlist headers separately from item pages. Quantify assumptions before naming databases.
• Problem Statement & Context note 24: Ordered lists are the product primitive — playback is a separate service. Quantify assumptions before naming databases.
• Problem Statement & Context note 25: Filter unlicensed items before returning playlist payloads. Quantify assumptions before naming databases.
• Problem Statement & Context note 26: Cache playlist headers separately from item pages. Quantify assumptions before naming databases.
• Problem Statement & Context note 27: Ordered lists are the product primitive — playback is a separate service. Quantify assumptions before naming databases.
• Problem Statement & Context note 28: Filter unlicensed items before returning playlist payloads. Quantify assumptions before naming databases.

Why interviewers care

Playlist System interviews reward crisp scope, explicit trade-offs, and failure stories—not generic microservice diagrams.

Interview checkpoint

Name one failure story for Problem Statement & Context that proves you understand real outages, not happy-path diagrams.