Senior AI Data Engineer

Comply is looking for a Senior AI Data Engineer to turn our semantic data models into the knowledge graphs, vector search, and LLM-powered pipelines that make Comply’s financial and regulatory data genuinely AI-ready. Comply is the world’s leading aggregator of financial and regulatory data for compliance, ingesting and enriching vast volumes of complex, high-variance data from hundreds of brokers and data providers across the US and beyond. The Comply Data Platform is a strategic initiative at the heart of that mission: a modern, cloud-native semantic layer built from the ground up-using JSON-LD as its semantic language-to power AI-driven insights, regulatory analytics, and next-generation data products. In this hands-on role at the intersection of knowledge representation, AI infrastructure, and data platform engineering, you’ll own the delivery of semantic layer components, work hand-in-hand with data engineers, architects, and our ontologist, and make sure AI-ready data products are reliable, performant, and actually adopted. You’ll join a new team being created to enable Comply’s AI ambitions.

You’ll love this role if you’re an engineer who turns semantic and ontological concepts into concrete, production-grade systems-and you’re energized by greenfield environments where the architecture is still being defined.

What you’ll do

• Implement JSON-LD-based semantic models from our ontologist into production data systems, and build knowledge graph structures that reflect canonical domain models.

• Develop and manage graph database schemas, queries, and ingestion pipelines, keeping semantic consistency between ontology definitions and downstream data products.

• Design and implement embedding pipelines that represent Comply’s financial and regulatory data in vector space.

• Build and operate vector database infrastructure for semantic search and similarity retrieval.

• Implement RAG architectures that ground LLM outputs in Comply’s proprietary data, and evaluate and integrate LLM tooling suited to our use cases.

• Build reliable, observable data pipelines that feed the semantic layer from upstream broker and regulatory sources, applying DataOps practices such as testing, monitoring, lineage tracking, and SLAs.

• Work with Data and Backend Engineers to embed semantic models into APIs and data contracts, ensuring the semantic layer scales with data volume and platform growth.

• Partner closely with the Ontologist so implemented models faithfully reflect domain intent, and support consuming application teams in adopting AI-ready data products.

What success looks like in your first 90 days

• You’ve shipped your first semantic layer components into production, turning ontology models into working knowledge graph structures.

• You’ve stood up or extended embedding and vector infrastructure that a consuming team is actively using for semantic search or RAG.

• You’ve put DataOps practices in place so pipelines are observable and maintain semantic quality from ingestion through to consumption.

• Strong hands-on data engineering experience, with a focus on semantic or AI data infrastructure.

• Experience building and operating knowledge graphs or graph databases (e.g., Jena Fuseki, Neo4j, Amazon Neptune, or equivalent).

• Experience with vector databases and embedding pipelines (e.g., Pinecone, Weaviate, Qdrant, pgvector).

• Practical experience implementing RAG architectures or LLM-integrated data pipelines.

• Familiarity with semantic web standards-JSON-LD, RDF, OWL, or SKOS.

• Strong Python skills and experience with data pipeline frameworks.

• Experience with cloud-native data platforms (AWS, Azure, or GCP).

Nice-to-have

• Exposure to domain-driven design (DDD) and bounded contexts.

• Experience working directly with ontologists or knowledge engineers.

• Familiarity with data contracts and data product frameworks.

• Experience with DataOps tooling, data reliability, or data observability platforms.

• Background in financial services, RegTech, or compliance data.