Setting up a DGX Spark local inference cluster with 5 role-based Hermes agents
natesiggard · x · 2026-09-25
The author spent day one setting up DGX Spark and a local inference cluster, installing Hermes plus Actual (private local AI inference) on 5 machines, each with its own identity.
- A "family office" model: each Hermes agent got a distinct role, boundaries, and an inference preference cascade
- The agents built their own coordination webapp hosted on Cloudflare
- Actual installs with one curl command, supports macOS/Linux/Windows, pools idle machines' compute over an encrypted relay, and runs open models privately on your own hardware
More from coding & agent
- Why inference speed will be AI products' biggest competitive edge in 1–2 years — itsOmSarraf_ · 2026-11-11
- What happened when I moved vector retrieval into Snowflake SQL instead of orchestrating RAG outside — dima806_dima · 2026-11-11
- OpenAI engineer runs Codex for 26 days straight hunting Atlantis, spawning "vibe archaeology" — Polymarket · 2026-11-11
- He replaced a $5,000/month ghostwriter with 11 Claude Skills and earned $47,000 from one post — aryanXmahajan · 2026-11-11
- Refero Styles ships 2,000+ DESIGN.md files to give Claude Code real design taste — bbssppllvv · 2026-11-11
- Agent circuit breaker: openclaw shorts bot halts itself after 3 failed experiments — Rich_Standard_1724 · 2026-11-11