Knowledge Graph Data Engineer

Description

Job ID BT-473 for a Knowledge Graph Data Engineer.

The team is building the data and retrieval base for a Graph RAG application. It combines a property graph, vector search and database queries behind a LangGraph agent. The work turns material from several sources into dependable graph data and makes it reachable through retrieval APIs and MCP tools.

Responsibilities

• Write Python pipelines that take in, tidy, split into chunks, enrich and index both structured and unstructured data.
• Design entities and relationships in a property graph, recording where each fact came from and linking it back to its source.
• Create workflows that extract and match entities so records from different sources line up.
• Build embedding pipelines and keep vector indexes current for semantic search.
• Combine vector, keyword, graph traversal and database lookups into one retrieval approach.
• Build and connect MCP servers and tools that give a LangGraph agent access to vector search.
• Connect the Oracle MCP server so the agent can read relevant database data.
• Use graph algorithms to uncover links between items and improve retrieval.
• Measure graph quality, retrieval relevance, whether answers are grounded in sources, and pipeline reliability.
• Work with application engineers and stakeholders to shape use cases into graph models, retrieval flows and agent tools.

Requirements

• Solid computer science foundations: data structures, algorithms, software design and testing.
• Skilled in Python, with experience building maintainable data pipelines or backend services.
• Working knowledge of cloud basics: compute, storage, networking, security and deployment.
• Hands-on work with property graph modelling, queries, traversals and graph algorithms.
• Prior work with embeddings and vector indexes, covering semantic and hybrid search.
• Understanding of retrieval-augmented generation and how to judge the quality of retrieved context.
• Able to handle inconsistent or incomplete data and explain technical choices clearly.
• 10+ years of relevant experience.
• Bachelor's-level training in engineering, mathematics or computer science.
• Preferred: experience with Oracle Cloud Infrastructure, Oracle Database or Oracle Property Graph.
• Preferred: experience building agents with LangGraph or a similar framework.
• Preferred: experience building or integrating Model Context Protocol servers and tools.
• Preferred: familiarity with entity resolution, information extraction or knowledge graph construction.
• Preferred: experience with CI/CD, containers, infrastructure as code and production monitoring.
• Preferred: a master's degree.

Conditions

Do not apply without an active Poly clearance. Candidates without one will not be considered.