Job DescriptionSr Consultant-Knowledge Graph / RAG Agentic AI ExpertYou will join Dell, driving innovation at the intersection of knowledge graphs and Generative AI. This role focuses on graph-based modeling and reasoning as well as GenAI, LLMs, and agentic workflows-delivering intelligent, explainable, and scalable solutions for Dell Services and platforms. You will advance the state of the art in graph technologies, and LLM/multi-modal integration.
We work across research and engineering-partnering with leading academics, industry experts, and world-class teams-to advance methodologies, tools, and evaluation practices. Our mission is to combine symbolic knowledge with statistical learning to deliver resilient AI that retrieves, reasons, and acts with confidence at scale.
Join us to do the best work of your career as a
Sr Consultant-Knowledge Graph / RAG Agentic AI Expert and make a profound social impact our
Data Science Team in
Austin, Texas.
What you'll achieveYou will define and operationalize the semantic architecture-taxonomies, ontologies, and knowledge graphs-that enables autonomous, agentic AI workflows across Dell Services. You will translate complex data into actionable decisions by grounding LLM/RAG systems in governed knowledge, designing robust evaluation and observability, and collaborating with leaders and engineers to drive measurable business outcomes.
You will:- Define end-to-end architecture for LLM, RAG/GraphRAG, and multi-agent systems, including data pipelines, deployment, observability, governance, and cost controls.
- Design ontologies and taxonomies; build and operate enterprise knowledge graphs (Neo4j, RDF/OWL), integrating structured, semi-structured, and unstructured sources with lineage and scalable Cypher/SPARQL queries.
- Develop extraction and linking pipelines for entities and relations, including disambiguation, conflation, deduplication, canonicalization, and quality assurance.
- Build production LLM and agentic workflows (e.g., LangGraph, LlamaIndex) for KG enrichment and natural-language-to-graph query generation with safe tool use, tracing, and human-in-the-loop where needed.
- Implement advanced retrieval that blends vector search, symbolic reasoning, and KG retrieval, including GraphRAG, hybrid dense/sparse retrieval, ontology-guided search, and contextual agents.
- Establish evaluation and observability using OpenTelemetry, SLIs/SLOs, and metrics for RAG/GraphRAG/graphs-such as faithfulness, grounding, multi-hop accuracy, entity-resolution precision/recall/F1, link-prediction MRR/Hits[redacted], schema/SHACL validation rates, and query latency; lead metadata governance, audits, drift detection, and remediation with cross-functional teams.
Take the first step towards your dream careerEvery Dell Technologies team member brings something unique to the table. Here's what we are looking for with this role:Essential Requirements- Deep expertise in taxonomy, ontology, and semantic modeling with hands-on experience building and operating enterprise knowledge graphs; fluency in Cypher and SPARQL.
- Proven delivery of production LLM, RAG/GraphRAG, and multi-agent systems with guardrails, safe tool use, tracing, and lifecycle management using frameworks such as LangGraph and LlamaIndex.
- Strong Python and AI/ML skills with practical NLP for extraction and normalization, plus rigorous experiment design, error analysis, and A/B testing.
- Knowledge of graph ML and retrieval including graph embeddings and algorithms, hybrid text-plus-graph retrieval, and reranking, and multi-hop reasoning.
- Clear communication and leadership in agile environments with the ability to influence product direction, mentor engineers, and engage technical and non-technical stakeholders.
- Experience establishing evaluation and governance for RAG/GraphRAG and graphs.
Desirable Requirements- Bachelor's degree with 12+ years of industry experience, or Master's degree with 10+ years, or equivalent experience.
- Familiarity with cloud platforms, fine-tuning (LoRA/QLoRA), RLHF/DPO, GPU inference stacks (vLLM, TensorRT-LLM), and ultra-low-latency, high-throughput serving also experience with metadata governance and policy-as-code, AI governance and LLM security (e.g., OWASP GenAI/LLM Top 10), red-teaming and post-market monitoring
