Senior Agentic AI Engineer (VP)

About Citi

Citi, the leading global bank, has approximately 200 million customer accounts and does business in more than 160 countries and jurisdictions. Citi provides consumers, corporations, governments, and institutions with a broad range of financial products and services, including consumer banking and credit, corporate and investment banking, securities brokerage, transaction services, and wealth management.

As a bank with a brain and a soul, Citi creates economic value that is systemically responsible and in our clients' best interests. As a financial institution that touches every region of the world and every sector that shapes your daily life, our Enterprise Operations & Technology teams are charged with a mission that rivals any large tech company. Our technology solutions are the foundations of everything we do from keeping the bank safe, managing global resources, and providing the technical tools our workers need to be successful to designing our digital architecture and ensuring our platforms provide a first-class customer experience. We reimagine client and partner experiences to deliver excellence through secure, reliable, and efficient services.

Our commitment to diversity includes a workforce that represents the clients we serve from all walks of life, backgrounds, and origins. We foster an environment where the best people want to work. We value and demand respect for others, promote individuals based on merit, and ensure opportunities for personal development are widely available to all. Ideal candidates are innovators with well-rounded backgrounds who bring their authentic selves to work and complement our culture of delivering results with pride. If you are a problem solver who seeks passion in your work, come join us. We'll enable growth and progress together.

About the Team

The Senior Agentic AI Engineer is a high-impact technical professional who sits at the convergence of advanced AI engineering and enterprise-scale financial technology — a role designed for those who do not just follow the frontier of generative and agentic AI, but actively shape it. This is a position of real consequence: you will help define how one of the world's most complex financial institutions architects, governs, and deploys AI systems that are grounded, reliable, and built to operate at scale. We are looking for a practitioner of deep technical conviction — someone who brings mastery of context engineering, retrieval systems, knowledge graphs, and multi-agent orchestration, and who is energized by the challenge of making these capabilities production-ready within the rigorous demands of a regulated, global enterprise.

The successful candidate will operate at the center of a cross-functional ecosystem — partnering with AI architects, engineering leads, and business stakeholders to design and deliver agentic AI solutions that meaningfully advance Citi's automation and operational efficiency agenda. You will architect sophisticated agent systems, mentor the next generation of AI engineers, and contribute to a culture of technical excellence that sets the standard for how AI is built and governed across the organization. Your work will directly strengthen Citi's Controls Technology platform, and the ripple effects of what you build will be felt across teams, products, and the millions of clients and communities Citi serves every day.

Responsibilities

Collaborate with AI architects, leads, and stakeholders to design and implement generative and agentic AI solutions that address business challenges.

Architect advanced context engineering strategies — context layering, chaining, compression, pruning/offloading, and memory management — to maximize reliability, provenance, and token efficiency in production.

Design and implement advanced generative AI methods, including sophisticated prompt engineering and Retrieval-Augmented Generation (RAG) .

Build and optimize RAG systems , including hybrid search, multi-vector retrieval, and re-ranking pipelines.

Design and implement knowledge graphs and Graph RAG architectures to enable multi-hop reasoning, explainability, and traceable, grounded responses for high-value business domains.

Architect agentic workflows and multi-agent systems using Google Agent Development Kit (ADK) and comparable frameworks (LangGraph, Microsoft Agent Framework, CrewAI), applying orchestration patterns such as supervisor/worker, hierarchical, and peer-to-peer.

Design robust agent harnesses — governance, constraints, feedback loops, state/session management, and execution controls that make long-running agent systems reliable and safe.

Integrate agents with tools and data via the Model Context Protocol (MCP) and orchestrate inter-agent collaboration and task delegation via the Agent2Agent (A2A) protocol.

Support the integration of GenAI and agentic applications into production environments, ensuring robust deployment, scalability, observability, and maintainability.

Contribute to the development and optimization of real-time and streaming AI solutions.

Stay current with the latest advances in generative and agentic AI and actively share knowledge with the team.

Ensure adherence to ethical AI guidelines, guardrails, agent isolation/sandboxing, data privacy, and compliance standards.

Mentor junior team members, provide code reviews, and foster a culture of technical excellence.

Qualifications

5–7 years of experience in AI/software development, including significant experience in Generative AI and agentic AI.

Demonstrated portfolio of successful AI-driven projects in a business environment.

Experience working with AWS (or equivalent) cloud infrastructure for AI/GenAI.

Required Technical Skills

Deep, hands-on expertise in core generative AI concepts — foundation models, LLMs, embeddings, tokenization, and context-window management.

Advanced skills in prompt engineering and context engineering , including familiarity with prompt design tools/frameworks and dynamic context orchestration.

Strong experience building RAG systems , including chunking strategies, hybrid search, and multi-vector retrieval.

Practical experience designing knowledge graphs and Graph RAG pipelines (e.g., using graph databases such as Neo4j or ArangoDB) for relationship-aware, multi-hop retrieval.

Proven experience building agentic AI systems with Google ADK and/or comparable frameworks (LangGraph, Microsoft Agent Framework, CrewAI, OpenAI Agents SDK), including tool/function calling, planning, and memory.

Strong grasp of multi-agent orchestration patterns (supervisor/worker, hierarchical, peer-to-peer) and harness engineering (governance, feedback loops, execution controls, agent isolation/sandboxing).

Hands-on experience with agent interoperability protocols — the Model Context Protocol (MCP) for tool/data access and the Agent2Agent (A2A) protocol for inter-agent collaboration.

Experience with agent observability and evaluation (e.g., tracing, OpenTelemetry-based tooling) for production agent systems.

Proficiency with major GenAI APIs (OpenAI, Gemini, Claude, etc.) and orchestration frameworks such as LangChain and LlamaIndex .

Strong skills in NLP (NER, dependency parsing, text classification, topic modeling).

Proficiency with vector databases and embedding models for large-scale retrieval.

Experience with containerization ( Docker ), orchestration ( Kubernetes ), and CI/CD pipelines for AI/agentic applications.

Solid understanding of AI compliance, guardrails, and responsible AI practices.

Strong skills in Python and experience with data preprocessing, document ingestion, and API development.

Required Soft Skills

Strong collaboration skills to work effectively in cross-functional teams.

Analytical and proactive approach to problem-solving.

Clear communication skills for both technical and non-technical audiences.

Eagerness to learn, innovate, and mentor