Categories › AI & Agents › Models, LLM APIs, routing, serving, training & GPUs
Engineering log of self-hosted AI on NVIDIA DGX Spark (GB10/SM121A). 60+ articles indexed.
Remote server: your requests and data go to mcp.sovgrid.org. Claude Code: claude mcp add --transport http self-hosted-ai https://mcp.sovgrid.org/self-hosted-ai?ref=mcp-io · Claude app: Settings → Connectors → Add custom connector.
The tools this server listed when we connected to it, in its own words.
search_blog — Search the Sovereign AI Blog for articles matching a natural language query, optionally filtered by tag and sorted by relevance or date. Behlist_articles — List all blog articles. No TF-IDF computation — pure database listing. Use to browse the full corpus, paginate through articles, or filter bget_article — Retrieve the full content of a blog article by its slug. Returns the article body (Markdown) plus metadata. If the slug does not match any adiagnose_sglang — Validate an SGLang configuration for NVIDIA DGX Spark (GB10/SM121A). Pure pattern-matching against known failure modes documented in the Sovlist_tags — List all topic tags used across the Sovereign AI Blog corpus, with article counts. Use this to browse the topic space before calling search_Not tested by KitScout yet. When our harness runs it, pros, cons and the tasks we tried appear here, with the date and evidence.