Join EY's Cyber Analytics and Data Science team as a Sr. Agentic AI Engineer. This pivotal role involves architecting, developing, and deploying sophisticated agentic AI systems, multi-agent frameworks, and intelligent automation solutions to elevate our cybersecurity capabilities. You will leverage cutting-edge machine learning, LLM engineering, and reasoning systems to tackle large-scale challenges and pioneer the next wave of autonomous cyber-analytics.

Your responsibilities will span the entire lifecycle of agentic AI applications, ensuring they support predictive, prescriptive, and autonomous cybersecurity operations. You'll collaborate closely with stakeholders across Cybersecurity, IT, and Data Engineering to accelerate the adoption of multi-agent systems, RAG-enhanced intelligence, and enterprise AI platforms.

This is a hands-on position for a technologist and problem-solver with deep expertise in emerging AI technologies and a passion for building robust, scalable autonomous systems.

Architect, design, and deploy agentic AI workflows using frameworks like LangChain, LangGraph, and AutoGen. Build multi-agent systems capable of autonomous reasoning, planning, and collaboration within cybersecurity. Implement advanced agent coordination strategies and design secure, scalable Agent Development Kit (ADK) pipelines. Develop Retrieval-Augmented Generation (RAG) pipelines for seamless knowledge integration and optimize them for high-accuracy decision support. Fine-tune and engineer Large Language Models (LLMs) for specialized cybersecurity tasks, integrating external APIs and custom models. Lead the development of an enterprise-grade AI platform for orchestrating LLMs, RAG, vector databases, and multi-agent protocols. Implement MLOps practices, including CI/CD, versioning, and lifecycle management for autonomous systems. Establish robust monitoring and observability for agentic system behavior and automate cybersecurity processes. Extract, transform, and analyze data from diverse cybersecurity sources, applying ML for anomaly detection and pattern recognition. Collaborate with cybersecurity SMEs to identify automation opportunities and clearly communicate complex AI concepts to various audiences.

A minimum of 5 years in software development, AI/ML engineering, or data science is required.

Essential experience includes at least 1 year in Cybersecurity, with exposure to IAM (SailPoint, CyberArk) and SIEM/SOAR (Splunk, QRadar).

You must have at least 1 year of hands-on experience building agentic AI or multi-agent applications, including LLM-driven workflows.

Proficiency in Python and working knowledge of SQL are mandatory.

Direct experience with LLM/SLM APIs, embeddings, vector databases, RAG architecture, and memory systems is crucial.

Experience deploying AI workloads on GCP (Vertex AI) and IBM WatsonX is expected.

Familiarity with agentic AI protocols, ADKs, LangGraph, AutoGen, or similar orchestration tools is necessary.

Practical implementation experience with Model Context Protocol (MCP) for agent-level context management is required.

At least 1 year of experience with frameworks like LangChain, LlamaIndex, OpenAI, Cohere, or Anthropic is essential.

Preferred qualifications include 2+ years in automation or RPA, 2+ years building on AWS, experience with data visualization platforms, and 2+ years working with APIs and microservices.