Llm Knowledge Graph Jobs in Georgia (NOW HIRING)

Job title: Senior Knowledge Graph Engineer

At The Coca-Cola Company, our vision is to craft brands and choices people love while refreshing the world in more ways than ever before. Data and intelligence are at the heart of how we understand consumers, accelerate growth, and create a more connected enterprise. The Senior Knowledge Graph Engineer play a pivotal role in building the semantic foundations that drive consistent, trusted, and actionable data across our global system. This role will be part of a forward-looking Data Engineering and Platforms team, enabling scalable use of trusted data, advanced analytics, and knowledge graphs to power decision-making.

Semantic clarity is essential for interoperability across markets, AI models, and platforms. This role will drive the creation of a governed semantic and graph foundation that connects fragmented data sources and enables agents, copilots, analytics, and operational decision-making. This is an individual contributor role focused on hands-on technical leadership, solution design, and delivery excellence rather than direct people management.

Core Responsibilities

• Lead the design, development, and maintenance of enterprise ontologies, taxonomies, controlled vocabularies, and graph models to enable semantic consistency and interoperability.

• Define modeling standards, reusable patterns, and implementation strategies for ontologies, entity relationships, upper ontology concepts, and property graph structures.

• Integrate graph solutions with enterprise data stores, APIs, MCP servers, and related technologies to meet stakeholder needs.

• Architect scalable mapping pipelines that connect distributed physical data sources to the logical graph layer without data redundancy.

• Enable AI and machine learning through structured knowledge representations that improve inference, entity resolution, and data discoverability.

• Use LLMs, GenAI, rules engines, reusable frameworks, and automation utilities to curate, build, adapt, and evolve the corporate ontology catalog.

• Implement semantic validation, formal reasoning, and performance monitoring frameworks to ensure model correctness, scalability, auditability and reliability.
  

• Design semantic layers that explicitly bind underlying physical data tables to the enterprise ontology, ensuring autonomous agents and subagents are grounded in deterministic business logic rather than probabilistic LLM outputs.
  

• Develop context-injection and semantic routing patterns that allow multi-agent systems to securely query and traverse the knowledge graph for complex, multistep reasoning and planning.
  

• Establish the graph model as the foundational long-term memory and context engine for enterprise copilots, enabling agents to maintain state and context across disjointed user sessions.

Required Qualifications & Experience

• Bachelor's or master'sdegree in information science, library science, ontology, semantics, computational linguistics, computer science, or related field.

• 2+ years of experience defining and implementing production-grade knowledge graphs and ontologies.

• Ability to develop and implement ontologies and data models in collaboration with stakeholders across data management, search, product management, machine learning, and other enterprise initiatives.

• 3+ years of experience with knowledge graph technologies such as RDF, OWL, SHACL, SKOS, LPG, and SPARQL.

• At least 2 years of experience or training with ontology and linked data tools such as Protege, TopQuadrant, Stardog, Jena, or Data.World.

• Expert proficiency with graph query languages such as Cypher, GQL, or SPARQL.

• Familiarity with enterprise ontology management suites and governance frameworks, including Knowledge Graph (organizational, GraphRAG, Query/Traversal) patterns.
  

• Hands-on experience integrating knowledge graphs with LLM orchestration and agent frameworks (e.g., LangChain, AutoGen, Semantic Kernel) to build productiongrade GraphRAG pipelines.
  

• Proficiency in hybrid retrieval strategies, combining vector embeddings with graph traversals to optimize agent context windows.

Preferred Qualifications

• Understanding of the development of ontologies and the use of controlled vocabularies and thesauri in enhancing the discovery of management of enterprise data.

• Evaluate the current state of semantic pilots, data assets, and structural mappings across the organization.

• Standardize the core taxonomical conventions and architectural blueprints for initial multi-domain integration.

• Demonstrate a measurable reduction in AI hallucination rates and a quantifiableincrease in autonomous multi-step task completion by leveraging the governedsemantic foundation.

• Showcase a quantifiable increase in context-retrieval accuracy and performance for dependent enterprise AI applications.