VP, Forward-Deployed AI Engineer - Production Agents

Citigroup Inc.; New York (NY); On-site; USD 142,000 - 213,000; Full time

Job summary

Citigroup Inc. in New York is seeking a Senior Forward Deployed Engineer to design and implement production-grade agent reasoning loops and resilient execution pipelines embedded with enterprise systems.

You will own the full engineering lifecycle from architecture to deployment, collaborating with business and technology teams to deliver scalable AI-powered solutions. You will work with LLM APIs, build MCP servers, ensure robust error handling, and contribute to automated tests and deployment

Qualifications

• 6+ years of software engineering experience with Python
• Hands-on with LLM APIs and enterprise integration
• Experience REST/gRPC, Docker, Git, Linux, CI/CD
• Understanding of prompt engineering and JSON output validation
• Experience in agile squads

Responsibilities

• Writes production-quality agent reasoning logic and sub-agent workflows in Python
• Develops MCP servers connecting AI agents to enterprise systems
• Implements resilient error handling, retries, HITL escalation
• Integrates guardrails for prompt-injection defense and PII filtering
• Collaborates with engineers to build regression testbeds and deployment harnesses
• Mentors junior developers and participates in sprint delivery and code reviews
• Adheres to Citi policies and compliance throughout the engineering process

Skills

Python

LLM APIs

REST/gRPC API development

Agile methodologies

Prompt engineering

JSON validation (Pydantic)

Education

Bachelor's degree or equivalent

Master's degree preferred

Tools

Docker

Git

Linux

CI/CD pipelines

Job description

Citigroup Inc. in New York is seeking a Senior Forward Deployed Engineer to design and implement production-grade agent reasoning loops and resilient execution pipelines embedded with enterprise systems.

You will own the full engineering lifecycle from architecture to deployment, collaborating with business and technology teams to deliver scalable AI-powered solutions. You will work with LLM APIs, build MCP servers, ensure robust error handling, and contribute to automated tests and deployment