Graph Engineer

About Neuron Factory

NeuronFactory is building an AI-native platform for pre-construction for large-scale, multibillion dollar construction companies around the world. We help estimators, project managers, risk teams, lawyers, and the many other professionals working in pre-con analyze tenders, generate RFQs, compare supplier quotes, and create comprehensive project estimates - often transforming tedious multiday day manual processes into an intelligent, collaborative experience.

We're at an exciting inflection point: our traction is strong, and now we need to transform how our client’s teams collaborate on these complex workflows.

The Role

We're looking for a Graph Engineer to architect and build the knowledge graph that powers our platform’s understanding of complex construction projects. This isn't about plugging documents into a vector database—it's about designing a graph structure that captures the intricate relationships between tenders, specifications, drawings, RFQs, quotes, subcontractors, and regulatory requirements. You'll work at the intersection of knowledge representation and practical AI systems, turning terabytes of unstructured construction documentation into a queryable knowledge structure that enables our agents to reason about projects the way experienced estimators do.

What You'll Do

What We're Looking For

Required

• ‍Graph-first thinking: You've designed a knowledge graph from first principles in a complex, real-world domain (legal, medical, financial, engineering, or similar). You understand that graph modeling is fundamentally different from relational or document modeling.
• ‍Deep graph database experience: Production experience with Neo4j or comparable systems (Amazon Neptune, TigerGraph, JanusGraph). You think naturally about Cypher/Gremlin query optimization, index strategies, and memory management.
• ‍Ontology design: Experience creating formal ontologies or taxonomies. You understand the tradeoffs between expressiveness and queryability, and you know when to normalize vs. denormalize for performance.
• ‍RAG foundations: You understand vector embeddings, semantic search, and retrieval-augmented generation. More importantly, you understand where RAG falls short and why graph-based approaches matter for complex reasoning.
• ‍Strong engineering fundamentals: Python fluency. Experience building data pipelines that process diverse document types at scale. Comfort with cloudinfrastructure (AWS preferred).

Preferred

• Experience in construction, engineering, or another domain with complex document interdependencies
• Background in NLP/NER for entity extraction from unstructured text
• Familiarity with LLM-based information extraction and structured output
• Experience with graph neural networks or knowledge graph embeddings
• Contributions to open-source graph tools or published work on knowledge graphs

Technical Environment