Knowledge Graph Engineer (Team Lead)

Cedar•San Diego, CA

11h•$180,000 - $250,000•Remote

About The Position

Senior Knowledge Graph Engineer (Team Lead) Help reinvent how housing gets designed, permitted, and built in America. Cedar builds software that accelerates urban housing delivery — from land acquisition through approved designs and construction. Our systems have cut project development timelines by months across active markets, and we're scaling to more cities. Our team blends architects, engineers, and product builders. We design real buildings — and we build the software systems that make great architecture scalable. The Role Cedar is hiring a Senior Knowledge Graph Engineer to lead the team that transforms America's regulatory landscape into structured, queryable knowledge. Zoning ordinances, land development codes, and entitlement requirements are the binding constraints on every housing project — and they're buried in thousands of pages of semi-structured legal text that varies by city, county, and state. Your team turns that text into structured, evaluable data that powers Cedar's products. This is a leadership role. You'll own the technical direction of a ~4-person engineering team responsible for Cedar's core knowledge layer — the data infrastructure that every other team builds on. You'll set the architecture, define the schema, ship the query APIs, and grow the team. If you've been a senior IC at a larger company and want to own a domain end-to-end, this is that opportunity.

Requirements

Knowledge representation and graph-based data modeling. You've built systems that encode complex, real-world rules as queryable structured data — knowledge graphs, ontologies, rule engines, or similar.
Document processing and structured extraction. Production experience turning unstructured or semi-structured text (PDFs, legal/regulatory documents, or similar) into clean, versioned structured data.
Rule engine or constraint system design. Experience with expression trees, decision tables, DSLs, or similar approaches to representing and evaluating conditional logic.
Data engineering fundamentals. ETL pipelines, data quality monitoring, versioning — the unglamorous work of keeping structured data reliable as it scales.
Technical leadership. You've led or are ready to lead a small team. You can set architectural direction, define data contracts across teams, and mentor engineers.
Pragmatic builder mindset. You ship systems that work in production, not research prototypes. You iterate toward reliability with imperfect, real-world data.

Nice To Haves

Experience with regulatory, legal, or policy data
Familiarity with GIS, parcel data, or spatial datasets
Background in NLP, information extraction, or computational linguistics (spaCy, custom NER, etc.)
Experience with LLM-based structured extraction workflows
Prior work on applied AI in operational products
Startup experience or comfort in fast-moving, ambiguous environments

Responsibilities

Lead a team of ~4 engineers building Cedar's regulatory knowledge graph and the extraction, structuring, and query systems around it
Design rule schemas and knowledge representations that capture real-world regulatory complexity — conditional logic, cross-references, overlapping constraints — in a form that's programmatically evaluable
Build and operate extraction pipelines that ingest municipal zoning codes and regulatory documents, turning unstructured legal text into clean, versioned structured data
Own the query APIs and rule evaluation layer that other teams consume — resolving applicable constraints for a given parcel or project, with citations back to source text
Architect the system to scale across jurisdictions without linear effort, using a mix of NLP, LLM-assisted extraction, and human-in-the-loop QA workflows
Define how Cedar's AI agents consume structured regulatory knowledge as context
Set technical standards, mentor engineers, and make build-vs-buy decisions for your domain