01 / AI-first

2025 · Confidential · Prototype

Industrial Intelligence

Institutional memory for heavy industry: ingest maintenance documents, build a knowledge graph of assets, people, incidents, and parts, then answer field questions with citations.

Confidential

IndustrialKnowledge GraphAgentsNDA
Industrial Intelligence, shipped client work
  • 01

    Five-agent ingest-to-graph pipeline

  • 02

    Deterministic single-point-of-failure expertise scoring

  • 03

    Cited answers over plant documents

  • 04

    Interactive knowledge graph plus Risk Radar

The brief

When a senior plant engineer retires or leaves, the institutional knowledge of which machine has which failure history, which procedure applies, and who is the only person who can fix a specific asset leaves with them. That knowledge usually lives in scattered documents and one person's memory, not in a searchable system.

The outcome

  1. 01Answers "if this engineer retires tomorrow, which machines lose their only expert?" deterministically, not as a best-guess LLM answer
  2. 02Full document-to-graph pipeline: PDF/DOCX/XLSX ingestion → entity extraction → graph linking, in one flow
  3. 03Built and demoed end-to-end in a hackathon timebox on a five-agent tool-use loop

The build

A five-agent pipeline (ingest, entity-link, jargon-normalize, graph-traversal, risk-scoring) turns uploaded maintenance documents into a live knowledge graph of Assets, People, Incidents, Documents, Procedures, and Parts.

On top of the graph sits a Risk Radar that flags machines that depend on a single expert, and an Expert Copilot that answers plant-floor questions with citations back to the source document.

Specs

Platform
Web, chat interface + interactive knowledge graph + Risk Radar panel
Graph engine
In-memory graph (graphology), persisted per plant
Agents
Ingestor, Linker, Jargon-normalizer, Traversal, Risk-scorer
Ingestion
PDF, DOCX, XLSX parsing → entity extraction → graph linking

Stack

AI / ML
Anthropic ClaudeAgentic tool-use loop
Backend
Node.jsExpressTypeScriptMongoDB
Frontend
Next.js 15React Flow

Building something in this layer?

Start an inquiry

More AI-first

All AI-first