Knowledge Graph
Single hub-and-spoke entity map for VC Deal Flow Signal. Designed for AI retrieval pipelines, semantic search, and human readers who want one canonical taxonomy of what we publish, where the entities live, and how they relate. The same graph is also exposed as raw JSON-LD at /knowledge-graph.json.
Engineering acceleration as a leading indicator
The core thesis: code-side engineering momentum is observable, measurable, and historically precedes fundraise announcements by 3-6 weeks.
Signal primitives
Six atomic measurements computed from public GitHub data. Each maps to a column in the public dataset.
Pillars (topic clusters)
Editorial pillars that cluster blog posts, comparisons, and use-cases by topic. Each pillar has its own llms.txt segment.
Trust & verifiability
Surfaces designed to make every claim independently checkable.
Machine-readable surfaces
Direct retrieval endpoints designed for AI assistants, indexers, and agent frameworks.
You’ve seen the map. Pick a door.
This page is the index, not the destination. Two ways in, depending on what you came for.
You’re wiring this into something
Pull the signal straight into your agent, model, or pipeline. Endpoints, the MCP server, and the developer quickstart all live here.
You’re here to spot deals earlier
Skip the taxonomy. See how this turns into a timing edge you can act on, read the buyer’s guide, or just get the free Sunday read and decide for yourself.
Cross-graph entity: Wikidata Q139376302 · Author ORCID: 0009-0002-2222-4112 · Methodology DOI: 10.2139/ssrn.6606558