AI/ML Data Knowledge Graph Engineer

Build pipelines that extract knowledge from documents, systems, and community sources into the graph.

Turn unstructured and semi-structured content into structured knowledge objects.

Keep the graph current as a community’s knowledge changes.

Resolve entities, deduplicate, and connect knowledge across fragmented sources.

Enforce quality so members can trust what the graph tells them.

Detect and handle conflicts and gaps in the knowledge.

Preserve provenance so every piece of knowledge can be traced to its source.

Build the structure that lets the platform show its work and earn member trust.

Protect sensitive community knowledge with correct access and governance.

Partner with neuro-symbolic and applied AI to serve the graph into reasoning and retrieval.

Shape the graph so it supports both symbolic reasoning and neural retrieval.

Make knowledge access fast enough for production answers.

Build the data pipelines and platform the knowledge layer depends on.

Instrument the pipelines so quality and freshness can be measured.

Turn recurring ingestion needs into reusable connectors.

Measure the quality, coverage, and freshness of the graph against what communities need.

Build the evaluation that tells whether the knowledge layer is improving.

Use evidence to steer where to invest next.

You use AI to build the knowledge layer, from extraction and entity resolution to schema suggestions, while you own the correctness, structure, and trustworthiness of the graph.

The standard is human in partnership: AI accelerates the work, you own the judgment, the interpretation, and the call. The people who create the most value here are not the ones producing the most output. They are the ones turning evidence into clear, durable decisions.

To keep the boundary clear:

This is not a reasoning-architecture role. You build the graph the COGENT architecture reasons over; neuro-symbolic AI owns the reasoning.

This is not a general analytics role. You engineer a production knowledge graph, not dashboards and reports.

This is not an ingestion-only role. You own structure, quality, provenance, and how knowledge serves reasoning.

This is not a best-effort role. The graph is a production foundation members’ trust depends on.

We measure this role on outcomes the team can see:

Connected knowledge. A community’s scattered expertise becomes a connected, queryable graph.

Trustworthy answers. Quality and provenance let members trust and trace what the platform tells them.

Fresh and current. The graph keeps pace with a community’s changing knowledge.

Reasoning-ready. The graph serves both symbolic reasoning and neural retrieval well.

Reusable ingestion. New sources come online faster because ingestion is reusable.

Measured quality. Coverage, quality, and freshness are measured and improving.

You name the real problem in the data before reaching for a structure.

You care about quality, provenance, and trust as much as coverage.

You build pipelines others can run and extend.

You measure the knowledge layer honestly.

You share reusable connectors and patterns.

Graph databases (for example Neo4j-class systems) and graph query languages (Cypher, SPARQL, or GQL).

Data pipeline and orchestration tools.

Entity resolution and data-quality tooling.

Vector databases and embedding models for hybrid retrieval.

Python and SQL as primary languages.

Cloud data platforms and storage.

Building and serving the KO graph into the COGENT architecture and MINERVA.

Prior data or knowledge graph engineering at a software or AI company.

Experience building knowledge structures from messy, real-world sources.

A track record of production data systems with quality and provenance.

Experience supporting reasoning or retrieval systems is a plus.

You work most closely with Neuro-Symbolic AI, Applied AI, Data Engineering, and Platform Engineering. You build the KO graph that the COGENT architecture reasons over inside MINERVA.

We review every application, and we encourage you to apply even if you do not match every line above. Research shows that talented people, especially those from underrepresented communities, often hold back when they do not meet every qualification. If that is the only thing holding you back, apply anyway.

Sapience AI is an equal opportunity employer. We are committed to a workplace where everyone, regardless of background, has a voice in building what comes next.