Spotify Cuts Claude Code Token Costs 90% by Offloading File Reads to Cheaper Models

Spotify engineers built a Claude Code plugin that intercepts large file reads via hooks and routes them to a cheaper Gemini worker, cutting token usage by 90%. They argue most agent work is I/O rather than reasoning, and industry data suggests this consumption will keep growing toward 2028.

2026-09-07 ~ 2026-09-07 · 3 related posts

• Spotify engineer's Claude Code hook cuts token spend 90% by rerouting big file reads to Gemini — RexDouglass · 2026-09-07
• Spotify cut Claude Code token usage by 90% by routing big file reads to cheap models — aliscodes · 2026-09-07
• Spotify details the Portal mode setup that cut Claude Code token usage by 90% — aliscodes · 2026-09-07