Lead Python AI Engineer, Agentic Systems
Listed on 2026-08-30
-
Software Development
AI Engineer (Applied/Software), Backend Developer
Senior Python-AI Engineer, Agentic Systems
Citi is building some of the most consequential AI solutions in global financial services, and this Senior AI Engineer role sits at the heart of that mission — designing and developing production-grade Agentic AI systems for Funds Transfer Pricing and Financial Hedging platforms. You will take ownership of complex AI agent architectures from concept through to global-scale deployment, driving automation and decision intelligence across critical financial workflows.
This is an opportunity to shape the next generation of AI-powered financial technology at one of the world's leading institutions.
- Design and build multi-agent systems where autonomous agents collaborate to solve complex, real-world business problems across financial platforms.
- Develop and optimize Retrieval-Augmented Generation (RAG) architectures, including embedding strategies, vector databases, and retrieval pipelines, to improve the accuracy and reliability of AI-generated outputs.
- Implement planning and reasoning capabilities using knowledge graphs, rule-based reasoning, and search algorithms that enable agents to execute multi-step workflows reliably.
- Construct resilient agent architectures with built-in mechanisms for error recovery, self-correction, and feedback-driven improvement to maintain performance in production environments.
- Integrate large language models (LLMs), predictive models, and reasoning frameworks into agent systems that support complex financial decision-making workflows.
- Build APIs, tools, and microservices that connect AI capabilities with enterprise applications, enabling seamless integration at scale.
- Optimize AI systems for low latency and high throughput, applying techniques such as response streaming, caching, and token usage optimization to ensure performance and cost-effectiveness.
- Apply current AI research to real business challenges and collaborate with engineers across the team to elevate AI development practices and foster engineering excellence.
- Act as a subject matter expert, driving the technical strategy for AI within the Funds Transfer Pricing and Financial Hedging domains by staying abreast of state-of-the-art research and identifying opportunities for innovation.
- Mentor junior engineers on AI best practices and provide technical leadership across multiple project teams, fostering a culture of engineering excellence
- 7 or more years of professional experience in software development and system design, with a demonstrated history of delivering large-scale production systems.
- Proficiency in Python and SQL, with hands-on experience building production-quality Agentic AI solutions using frameworks such as Lang Chain, Llama Index, or equivalent tools.
- Practical experience developing multi-agent systems using frameworks such as Google ADK, Lang Graph, Auto Gen, or CrewAI, including implementation of planning, reasoning, and memory systems.
- Applied knowledge of large language models (LLMs) within agentic architectures, including designing and integrating APIs for AI services.
- Advanced skills in Prompt and Context Engineering, with the ability to structure, compress, and optimize information for consistent and high-quality model performance.
- Experience working with Vector Databases as part of AI retrieval and memory architectures.
Skills & Qualifications
- Experience working within the financial services industry, particularly in areas such as Funds Transfer, Pricing, Hedging, or Treasury technology.
- Proficiency in Java as a secondary programming language alongside Python.
- A Master's degree in Computer Science or a closely related field.
At Citi, you will work alongside talented engineers on technology that operates at global scale and drives real impact across financial markets. This is a high-visibility role within a firm-wide AI modernization initiative, offering the technical scope and career momentum that comes with building systems that matter.
- A hybrid working model with 3 days in the office and 2 days working remotely, giving you flexibility alongside meaningful in-person collaboration.
- The…
(If this job is in fact in your jurisdiction, then you may be using a Proxy or VPN to access this site, and to progress further, you should change your connectivity to another mobile device or PC).