Lead Agentic AI Engineer - FDE ( Forward Deploy Engineer)

Job description

Job Description

Lead Agentic AI Engineer — Forward Deployed Engineer (FDE)

• Role Overview & Mission As a Lead Agentic AI Engineer (Forward Deployed), you will operate at the intersection of applied GenAI engineering, autonomous multi-agent systems, and business domain problem-solving.

Unlike traditional back-office research roles, a Forward Deployed Engineer (FDE) is embedded directly with business stakeholders, product leads, and operations teams to translate complex, unstructured institutional workflows into high-impact, enterprise-grade Autonomous Agent solutions. You will spearhead the design, prototyping, evaluation, and end-to-end production deployment of multi-agent architectures, agentic tool orchestration (e.g., Model Context Protocol / MCP, tool-calling pipelines), dynamic RAG retrieval engines, and human-in-the-loop autonomous decision workflows across institutional platforms (e.g., KYC Intelligence, Funds Lifecycle, Document Processing, and Regulatory Compliance).

Key Responsibilities

• Architecture & Agentic Engineering Responsibilities: AI Solutioning & Translation: Formulate and define systems scope and objectives for complex AI/GenAI projects, fostering seamless communication and translating business requirements into technical specifications for Data Scientists and ML Engineers.

Implementation & Integration

Support the end-to-end AI implementation and system change processes—from translating Business Requirement Documents (BRDs) to functional specifications, through model deployment, testing, and integration with existing systems.

Customization & Scripting

Consult with business clients to determine functional specifications for AI applications, developing custom scripts, integrations, or workflows to tailor AI/GenAI products to the unique needs of the business.

Advanced Troubleshooting

Consult with users and clients to solve complex system and model integration issues through in-depth evaluation of business processes, AI/ML systems, and industry standards, recommending tailored technical solutions.

Feedback Loop & Product Advocacy

Act as a critical bridge between front-line business users and core AI research/engineering teams. Gather direct feedback, identify model performance pain points, and relay insights from the field to help guide the AI product roadmap.

Enablement & Training

Educate business and technical teams on how to leverage deployed AI solutions effectively, promoting best practices in prompt engineering, model interaction, and operational maintenance.

Risk & Impact Management

Identify and communicate risks and impacts (including AI model limitations, data privacy, bias, and ethical considerations), evaluating the business implications of applying GenAI/ML technologies to the current business environment.

Leadership & Collaboration

Act as advisor or coach to new or lower-level analysts and work as a team to achieve business objectives, performing other duties and functions as assigned.

Autonomy

Has the ability to operate with a limited level of direct supervision, exercising independence of judgment and autonomy while acting as an SME to senior stakeholders and/or other team members.

Compliance & Safeguarding

Appropriately assess risk when business decisions are made, demonstrating particular consideration for the firm's reputation and safeguarding Citigroup, its clients and assets, by driving compliance with applicable laws, rules and regulations, adhering to Policy, applying sound ethical judgment regarding personal behavior, conduct and business practices, and escalating, managing and reporting control issues with transparency.

Forward Deployed Leadership & Execution High-Velocity Prototyping to Production: Rapidly convert ambiguous business problems into functional, production-ready Proof of Concepts (PoCs) in 2–4 week cycles, followed by hardened enterprise deployment.

Client & Stakeholder Partnership

Sit alongside business heads, compliance officers, and operations leads to uncover automation bottlenecks and architect tailored AI agent interventions.

End-to-End System Integration

Partner with Java/Python backend, UI, and DevOps engineers to integrate agent services into core transactional systems, client lifecycle management (CLM) platforms, and CI/CD pipelines.

Governance & Model Risk Management (MRM)

Lead the technical defense and documentation of agentic workflows through enterprise Model Risk Management (MRM), Responsible AI, and InfoSec approval gates.

Required Qualifications & Experience

• Technical Expertise Experience: 7+ years of software engineering experience, with at least 3+ years dedicated to applied NLP, GenAI, LLM application development, and autonomous agent orchestration.

Programming Mastery

Advanced proficiency in Python (FastAPI, Asyncio, Pydantic, LangChain/LangGraph) and strong working knowledge of Java / Spring Boot or modern TypeScript/Node environments.

Agent Frameworks & Patterns

Proven experience building multi-agent systems, ReAct loops, Plan-and-Solve patterns, stateful workflows, memory management, and structured JSON output parsing.

Vector & Relational Databases

Hands-on experience with vector search engines (PostgreSQL with pgvector, Qdrant, Milvus, Pinecone) and relational/NoSQL datastores.

LLM APIs & Gateways

Deep familiarity with leading foundational models (Claude, GPT-4, Gemini) and enterprise gateways (e.g., R2D2, AWS Bedrock, Azure OpenAI) including token optimization, rate-limiting, and cost modeling.

Cloud & DevOps

Solid experience with containerization (Docker, Kubernetes/OpenShift), CI/CD pipelines, AWS/GCP cloud services (S3, IAM, Lambda/ECS), and enterprise observability (distributed tracing, OpenTelemetry). Soft Skills & Forward-Deployed Competencies Problem Solving Under Ambiguity: Ability to frame, scope, and deliver complex AI projects with minimal upfront requirements.

Executive Communication

Outstanding ability to articulate AI architectural trade-offs, confidence intervals, and ROI to technical and non-technical executive stakeholders.

Bias for Action

High-ownership mindset with the agility to shift from strategic architectural design to hands-on debugging in the same day.

• Preferred / Bonus Qualifications Prior experience in Institutional Banking, FinTech, Wealth Management, KYC/AML, or Fund Operations. Working knowledge of Model Context Protocol (MCP) or custom enterprise tool-calling frameworks.

Experience with Model Risk Governance (MRM), Fair Lending, or Regulatory AI Compliance frameworks. Contributions to open-source agentic AI libraries, research publications, or community AI toolkits ------------------------------------------------------ Job Family Group: Technology ------------------------------------------------------ Job Family: Applications Development ------------------------------------------------------ Time Type: Full time ------------------------------------------------------ Most Relevant Skills

Please see the requirements listed above. ------------------------------------------------------ Other Relevant Skills For complementary skills, please see above and/or contact the recruiter. ------------------------------------------------------ Citi is an equal opportunity employer, and qualified candidates will receive consideration without regard to their race, color, religion, sex, sexual orientation, gender identity, national origin, disability, status as a protected veteran, or any other characteristic protected by law.

If you are a person with a disability and need a reasonable accommodation to use our search tools and/or apply for a career opportunity review Accessibility at Citi . View Citi’s EEO Policy Statement and the Know Your Rights poster.