Principal Software Engineer

Realm; United States; On-site; USD 130,000 - 160,000; Full time

Job summary

Realm is looking for a technical leader to take ownership of the knowledge representation and retrieval layer in their AI-first software platform. This role involves setting the technical direction, making architectural decisions, and leading a growing engineering team. Ideal candidates will have production experience with knowledge graph systems like Neo4j, deep expertise in RAG architecture, and hands-on experience with vector databases. The position offers full remote work with 1-2 optional on-site visits per year.

Qualifications

• Production experience owning knowledge graph systems.
• Deep expertise in RAG architecture including retrieval and reranking.
• Hands-on experience with vector databases like Pinecone or Elasticsearch.

Responsibilities

• Own the knowledge representation and retrieval layer.
• Set technical direction and lead a growing engineering team.
• Drive key architectural decisions to production.

Skills

Knowledge graph systems ownership

RAG architecture expertise

Vector databases

Ontology design

Data fusion expertise

Familiarity with Apache Spark

Technical leadership

Iteration environments

Tools

Neo4j

Pinecone

Milvus

Elasticsearch

Job description

We are working with a profitable, established financial services company building an AI-first software platform used directly by banks and institutions to automate their operations.

With over a decade of client relationships and real-world financial data already on the platform, they are now building the knowledge layer from the ground up. This is a rare opportunity to own a greenfield system that is foundational to the intelligence of the entire platform, at a company that has the stability, clients, and resources to scale it properly.

The Role

You will take ownership of the knowledge representation and retrieval layer, validate the existing architecture, make the key technical decisions, and drive it through to production at scale.

This means owning the ontological structure of the knowledge graph, the data fusion layer across heterogeneous multi-format sources, and the full retrieval pipeline covering vector search, semantic ranking, and RAG. The system is the intelligence layer every agentic workflow on the platform depends on.

You will set the technical direction and lead a growing engineering team alongside the Director of AI.

Requirements

• Production experience owning knowledge graph systems, Neo4j or equivalent
• Deep expertise in RAG architecture including retrieval, reranking, and augmented generation pipelines
• Hands-on experience with vector databases, Pinecone, Milvus or Elasticsearch
• Strong command of ontology design, entity disambiguation, and navigable knowledge structures for agentic AI
• Experience with multi-source, multi-format data fusion across structured and unstructured data
• Familiarity with Apache Spark, Trino, Iceberg or Data Fusion
• Proven technical leadership and end-to-end ownership of architectural decisions
• Comfortable delivering in fast iteration environments without formal sprint structures
• Fully remote with 1-2 optional on-site visits per year