Principal Knowledge Graph Engineer

Wells Fargo & Company•Iselin, NJ

19h

About The Position

COO Technology delivers platforms for execution, resiliency, regulatory confidence, and continuous improvement across the Chief Operating Office, driving large-scale modernization by transforming complex operational data into actionable insight. This senior-level technical leader role involves owning ontology design and knowledge representation for the next generation of AI-enabled process intelligence. The individual will serve as the authority on how enterprise knowledge is structured, connected, and reasoned over, building the semantic foundation for AI systems to understand business operations. This work directly enables AI-driven process discovery, pattern recognition, process optimization, and automation at scale by unifying structured, semi-structured, and unstructured operational data into coherent, machine-interpretable knowledge graphs. The role requires partnering with engineering, process improvement, and analytics teams to translate real-world operational behavior (captured through clickstream data, videos, and transcripts) into formal models that align with documented processes, controls, and organizational structures. This position is ideal for someone who wants to shape how agentic AI, large language models, and knowledge graphs converge to unlock deep operational insight and drive measurable enterprise outcomes.

Requirements

7+ years of Engineering experience, or equivalent demonstrated through one or a combination of the following: work experience, training, military experience, education.
3+ years of experience in knowledge graph design, ontology engineering, or semantic technologies.
2+ years of experience working with LLMs and AI agents for information extraction, entity recognition, or knowledge base construction.

Nice To Haves

3+ years of hands-on experience with graph databases including schema design, query optimization, and large-scale data modeling.
Expertise in semantic web technologies (RDF, OWL, SPARQL) and their application to enterprise knowledge management.
Experience with agentic AI frameworks for automated knowledge extraction.
Proficiency in Python with experience in NLP libraries and frameworks.
Experience with process mining tools.
Background in operations research, business process optimization, or workflow analysis.
Experience in financial services, particularly in operations or risk management domains.
Knowledge of AI risk assessment, AI safety, and ethical considerations in enterprise AI.
Experience with cloud computing platforms.
Excellent communication skills with ability to translate complex technical concepts for non-technical stakeholders.
Advanced degree (M.S. or Ph.D.) in Computer Science, Information Science, or related field.

Responsibilities

Ontology Design & Architecture: Design comprehensive ontologies for knowledge graphs that capture: Unstructured data: procedure documents, SOPs, transcriptions of clickstream videos. Semi-structured data: process diagrams, BPMN XML files, diagrams extracted from PDFs. Structured data: Process metadata tables (applications, HR entities, management hierarchies, risk assessment units, processes, organizations, controls).
Agentic AI Prototyping: Use agentic AI frameworks to rapidly prototype and validate KG schemas in graph databases, enabling fast iteration on entity and relationship definitions.
LLM Prompt Engineering for Extraction: Define parameterized schemas and craft optimized prompts for LLMs to extract entities and relationships from unstructured and semi-structured data sources.
Cross-Functional Leadership: Serve as the ontology and schema expert across the Process Improvement workstream and guide the Analysis & Evaluation workstream on connecting observational data with process documentation.
Pattern Recognition: Enable the identification of similar patterns across processes and opportunities for high-impact AI automation
Knowledge Integration: Build frameworks that connect operational observations (from clickstream analysis) with documented processes (from SOPs and control documents).