Join EY and build a career defined by your unique contributions. Leverage our global scale, inclusive culture, and advanced technology to reach your full potential. Your distinct perspective is invaluable in shaping EY's future and creating a better working world for everyone. This role focuses on applying Artificial Intelligence to solve critical infrastructure challenges, including capacity planning, incident response, observability, automation, cost optimization, and resiliency.

Architect and implement AI/ML/LLM solutions within infrastructure operations, such as AIOps, self-healing systems, and predictive capacity planning. Design and develop agentic AI workflows for automated infrastructure tasks like ticketing, monitoring, and provisioning. Evaluate, fine-tune, and integrate various LLMs into enterprise infrastructure tooling. Build RAG pipelines and vector databases using infrastructure documentation. Write production-ready code, primarily in Python, for seamless integration with cloud platforms, ITSM tools, observability stacks, and CI/CD pipelines. Establish and maintain AI governance standards, ensuring data privacy, model security, and cost management. Collaborate with infrastructure leadership to define AI use cases and roadmaps, while also mentoring engineers on AI skills and leading the internal AI Center of Excellence for the infrastructure vertical. Oversee the entire lifecycle of AI initiatives, from Proof of Concept to production, incorporating MLOps/LLMOps practices.

A minimum of 10-12 years of experience in Infrastructure/Cloud engineering or architecture, with the last 3-4 years dedicated to applied AI/ML/LLM work. Proficiency in Python coding is essential, along with strong skills in API integration and SDK usage for tools like OpenAI, Anthropic, LangChain, and LlamaIndex. A thorough understanding of LLM principles, including prompting, fine-tuning, embeddings, RAG, and cost management, is required. Practical experience in building agentic AI systems using frameworks like LangGraph, AutoGen, or CrewAI, with demonstrated capability in tool/function calling, is crucial. Familiarity with vector databases such as Pinecone, Weaviate, FAISS, or Azure AI Search is necessary. A deep infrastructure background encompassing cloud architecture, networking, virtualization, ITSM, monitoring/observability, and automation using tools like Ansible and Terraform is expected. Experience with MLOps/LLMOps, including model deployment, monitoring, versioning, and cost governance, is also a key requirement. Strong architecture and solutioning capabilities, with the ability to design and articulate end-to-end systems, are vital.