Spegling

A personal authority system that turns your research into persistent, structured knowledge. Not an assistant. A mirror.

Spegling maintains continuity where other tools start from zero. It compiles your sources into a navigable knowledge graph, validates its own knowledge against reality, and makes your thinking visible and persistent across sessions. Every action flows through auditable authority chains.

How it works

  • Starts with thousands of architectural patterns and continuously updated research — then enriched by your projects, code, notes, and conversations into a large personalized knowledge corpus
  • Persistent memory across sessions. Context you would otherwise re-explain is carried forward with provenance and confidence tracking
  • Directed research campaigns: searches internal knowledge first, fills gaps from the web, writes articles, and reviews its own output through adversarial peer review
  • The system detects when its knowledge goes stale and refreshes it. Knowledge that doesn't maintain itself degrades silently — this doesn't

In practice

Used to develop turboquant-vllm, an open-source LLM compression library. Spegling conducted the research, identified approaches, and produced the benchmarks. Total GPU cost: $18. Full write-up at varjosoft.com/writing.

Interested in early access?

Spegling

Wiki

Loading articles...

Chat

News

Notes

    Patterns

    Repos

    Loading workspaces...
    -- articles -- pending agents
    ⌘K to open · ↑↓ to navigate · Enter to open · Esc to close