AI Solutions - Manager- Consulting - Location OPEN

EY • Hartford, DC

1d • $125,500 - $261,600 • Hybrid

About The Position

Our Artificial Intelligence and Data team helps apply cutting-edge technology and techniques to bring solutions to our clients. As part of that, you'll sit side-by-side with clients and diverse teams from EY to create a well-rounded approach to advising and solving challenging problems, some of which have not been solved before. No two days will be the same, and with constant research and development, you'll find yourself building knowledge that can be applied across a wide range of projects now, and in the future. You'll need to have a passion for continuous learning, stay ahead of the trends, and influence new ways of working so you can position solutions in the most relevant and innovative way for our clients. You can expect heavy client interaction in a fast-paced environment and the opportunity to develop your own career path for your unique skills and ambitions.
EY is investing significantly in our agentic AI platform, and the memory layer is the engineering foundation of that platform. Without memory, agents are amnesiacs they cannot accumulate context, learn from prior runs, or ground decisions in enterprise knowledge. We are hiring a Manager, AI/ML Engineer to build the memory layer end-to-end: the graph databases, ontologies, vector stores, retrieval services, and grounding pipelines that power our cognitive harness and the agents that run on top of it.
This is a hands-on engineering leadership role. You will architect and build the infrastructure — knowledge graphs, vector indices, hybrid retrieval, ontology services, memory APIs, and the integration glue that ties memory into the harness, agent runtimes, and downstream client solutions. You will work closely with our data-science leadership on memory representation and retrieval quality, and you will own the engineering delivery and production-readiness of the memory layer itself.

Requirements

• Strong AI/ML engineer who builds comfortable owning a memory service from schema to API to deploy, and writing the code that gets it there.
• Deep, hands-on knowledge of graph databases (Neo4j, Spanner Graph, Neptune, TigerGraph, or Stardog) — schema design, query languages (Cypher, GQL, SPARQL, Gremlin), and operating graph infrastructure in production.
• Strong grasp of ontologies and knowledge representation vocabulary, schema rules, instances, axioms, provenance and how ontologies support grounding and reasoning for AI agents.
• Solid working knowledge of cognitive harness / agent-runtime architectures memory, tools, policies, evaluation, observability and how memory infrastructure plugs into them.
• Hands-on experience with vector databases and hybrid retrieval — embeddings, ANN indexes, reranking, query rewriting, and semantic caching.
• Strong software engineering fundamentals Python (and ideally one of Java / Go / TypeScript), API design, testing, CI/CD, containerization, and observability.
• Experience designing and operating production AI systems on at least one major cloud (GCP, AWS, Azure, Databricks) including IAM, network controls, encryption, and responsible-AI guardrails.
• Track record of leading engineering teams setting technical direction, mentoring, and delivering production systems on time.
• Excellent communication skills able to explain memory and graph concepts to clients, and able to defend engineering decisions to architects, data scientists, and partners.
• Master's preferred in Computer Science, Software Engineering, Data Engineering, or a closely related technical field. Bachelor's with strong applied experience also considered.
• 6+ years of applied AI/ML or data engineering experience, with at least 2 years leading engineering teams or major workstreams.
• Demonstrable production experience with graph databases (Neo4j, Spanner Graph, Neptune, TigerGraph, Stardog, or similar), including schema design and query tuning.
• Hands-on experience designing and implementing ontologies and knowledge graphs for real systems — not just whiteboard.
• Production experience building retrieval / RAG / memory systems vector indices, hybrid retrieval, embedding pipelines, reranking, and grounding.
• Strong software engineering proficiency in Python and the modern AI/ML stack (PyTorch / TensorFlow, Hugging Face, LangChain / LangGraph, vector databases).
• Practical experience with at least one agent framework (Google ADK, Bedrock AgentCore, LangGraph, AutoGen, OpenAI Agents SDK) and one major cloud AI platform (Vertex AI, Bedrock, Azure AI Foundry, Databricks Mosaic AI).
• Experience operating in a client-facing or cross-functional environment and delivering production systems under enterprise constraints.

Nice To Haves

• Experience with semantic web standards (RDF, OWL, SHACL, SPARQL) and modern property-graph approaches in the same codebase.
• Experience designing memory architectures that meet regulated-industry constraints (SOX, HIPAA, GDPR) retention, audit, lineage, and explainability.
• Experience with neuro-symbolic patterns combining graphs and ontologies with LLM reasoning for higher-fidelity grounding.
• Experience with stream-based ingestion (Kafka, Pub/Sub, Kinesis) and CDC patterns into graph and vector stores.
• Open-source contributions to graph, retrieval, ontology, or agent-runtime projects.
• Prior consulting, product, or hyperscaler experience — comfortable in a fast, ambiguous environment with senior stakeholders.

Responsibilities

• Architect and build the memory layer of EY's cognitive harness end-to-end graph databases, vector stores, hybrid retrieval services, ontology services, memory APIs, and the integration into agent runtimes.
• Design and implement knowledge graphs and ontologies that support agent grounding vocabulary, schema rules, instances, provenance, and cross-domain mappings (sector and functional ontologies for Finance, Risk, Tax, Supply Chain, HR).
• Build and operate graph database infrastructure (Neo4j, Spanner Graph, Neptune, TigerGraph, or similar) schema design, ingestion, query optimization, and integration with the broader data platform.
• Engineer the vector and hybrid retrieval stack embeddings pipelines, vector indices (Vertex AI Vector Search, OpenSearch, Pinecone, Weaviate, pgvector), reranking, and lexical-plus-dense retrieval services.
• Build memory services for working, episodic, semantic, and procedural memory — including TTL, retention, consolidation, forgetting, and audit-grade provenance for regulated workloads (SOX, HIPAA, GDPR).
• Implement grounding pipelines that connect agent runtimes to the memory layer with low latency, citation tracking, and hallucination guardrails.
• Lead a team of AI/ML engineers and data engineers set technical standards, run code and design reviews, and mentor on production engineering rigor.
• Partner with data-science leadership on memory representation, retrieval quality, and evaluation, translating science prototypes into hardened production services.
• Stay abreast of AI and data trends new graph paradigms, embedding models, retrieval techniques, agent frameworks (Google ADK, Bedrock AgentCore, LangGraph) and recommend tools and patterns that fit our clients' existing ecosystems.
• Apply combined business and technical knowledge to develop and execute target memory architectures that enable implementation, monitoring, and ongoing evolution of agentic AI at scale.

Benefits

• medical and dental coverage
• pension and 401(k) plans
• a wide range of paid time off options
• Continuous learning
• Success as defined by you
• Transformative leadership
• Diverse and inclusive culture

What This Job Offers

• Job Type: Full-time
• Career Level: Manager