Spotify cut Claude Code token usage 90% with a two-model routing setup called Portal
SumitGup · x · 2026-09-06
Spotify's engineering team shared "Portal," the internal setup that cut their Claude Code token usage by 90%. The insight: most coding-assistant work (opening five files to answer a question about one, cloning patterns from neighboring tests) requires no frontier-level reasoning yet is billed at top rates.
Key mechanics:
- Two cheap "assistant" models: one opens files and returns short summaries, another writes repetitive code from examples straight to disk. The expensive model never sees any of it.
- Files over 350 lines are hard-blocked from the expensive model—not steered away, blocked.
- Rules-as-instructions failed first; engineers and the model routed around soft rules. Hard blocks at the routing layer worked where prompt-level guidance didn't.
Implication: most frontier-model usage in coding workflows doesn't actually need frontier reasoning—architecture-level routing beats prompt engineering.
Related event: Spotify Cuts Claude Code Token Usage by 90% with Dual-Model Routing(2 posts)→
More from coding & agent
- Ghost agent ran all night, cut per-turn cost from 20¢ to 4¢ with zero chatter — BLUECOW009 · 2026-09-07
- Astra + Blender + Dreamina Seedance 2.5 form a working AI video pipeline — eyishazyer · 2026-09-07
- One prompt runs a full 3D character pipeline in 30 minutes with GPT-6 Astra — _AustinCalvert_ · 2026-09-07
- Spotify engineer's Claude Code hook cuts token spend 90% by rerouting big file reads to Gemini — RexDouglass · 2026-09-07
- Bezalel launches: one MCP URL gives every agent memory, email, money and a computer — Rasmic · 2026-09-07
- Grok Coding Tip: Sending Repos to Cloud Agent Instead of Local Bot Saves the Weekly Quota — nikvassev · 2026-09-06