Running 100-200 agents daily: disk space is now the bottleneck, not compute
vincent_koc · x · 2026-09-12
Vincent Koc describes how his team's agentic engineering bottlenecks have evolved: first tokens, then compute, and now disk space — with electricity possibly next.
- The team runs 100-200 agents per day, but scaling further is hard without enough scratch space.
- Even hosting providers like Hetzner require manual requests for additional disk quota.
- Workarounds include using APFS virtual clones for pnpm packages so dependencies aren't fully copied into every worktree; however, agents tend to write to /tmp and random folders, making them hard to constrain.
Related event: Team Running 100-200 Agents Daily Hits Disk Space Bottleneck(3 posts)→
More from coding & agent
- GitHub Copilot app is building richer dashboard widgets for personalized data views — mariorod1 · 2026-10-31
- Todoist's self-healing agent auto-fixes 1 in 5 Sentry issues and opens PRs — zeeg · 2026-10-31
- Dev's Multica 1.0 build log: 432M tokens, 89 agent runs, just $183 — jiayuan_jy · 2026-10-31
- Stardock's QA now runs agents that play the game and file reports — draginol · 2026-10-31
- Limma MCP structures unstructured data end-to-end so agents never miss a detail — FlimsyIngenuity7248 · 2026-10-31
- Agent audit metric: can you name which tests ran and files changed without reading its summary? — ronbodkin · 2026-10-31