I think rest of it is pretty much expected from the role. We would not be requiring them to form new models but they need to have exposure to Gen AI integration patterns. I have just updated the title in the JD

 

-------------------------------------------------------------------------------------------------------------------------------

 

We are looking for an enthusiastic Python AI Engineer to join our Controls Technology team and support the development and integration of generative AI solutions. In this hands-on role, you will work under the guidance of senior developers and AI architects to help build retrieval-grounded, context-aware, and increasingly agentic AI applications. You will contribute to reliable, AI-driven features while growing your expertise across the modern GenAI stack. This role focuses on applying pre-trained and hosted foundation models — through context engineering, RAG, knowledge graphs, and agentic workflows — rather than training or fine-tuning models.

 

This is a growth-oriented role: you'll take ownership of well-scoped components, learn established patterns from senior engineers, and progressively increase your technical depth and independence.

 

Key Responsibilities

• Assist in building and integrating generative AI applications using pre-trained and hosted foundation models (via managed GenAI APIs and open-model endpoints).
• Support the implementation of context engineering workflows — assembling system instructions, retrieved knowledge, tool definitions, and conversation memory into reliable, token-efficient prompts, following established patterns.
• Contribute to prompt engineering (zero-shot, few-shot, chain-of-thought, role-based prompting) for AI-powered workflows.
• Help develop and maintain Retrieval-Augmented Generation (RAG) components, including chunking, embedding, and semantic/keyword search.
• Support the development of knowledge graph and Graph RAG pipelines under guidance to enable grounded, traceable responses.
• Contribute to agentic workflows — helping build AI agents with tool-calling and basic planning/memory, using frameworks such as Google Agent Development Kit (ADK), LangGraph, or CrewAI.
• Assist with integrating agents to external tools and data sources via the Model Context Protocol (MCP), with exposure to the Agent2Agent (A2A) protocol.
• Support the deployment, monitoring, and maintenance of GenAI and agentic applications in production environments.
• Perform data preprocessing, document ingestion, and basic API development for AI applications.
• Collaborate with data scientists and engineers to integrate AI capabilities into products.
• Participate in code reviews, testing, and documentation to ensure quality and reliability.
• Stay curious about advancements in GenAI and agentic AI, and share learnings with the team.

 

Required Technical Skills

• Proficiency in Python for GenAI development, data preprocessing, and scripting.
• Solid understanding of core generative AI concepts — foundation models, LLMs, tokenization, embeddings, and context windows.
• Hands-on experience with prompt engineering; foundational understanding of context engineering techniques.
• Practical experience (project or professional) building RAG systems, including chunking, vector databases, and semantic search.
• Familiarity with knowledge graphs and interest in Graph RAG for relationship-aware retrieval.
• Exposure to agentic AI development — building tool-using agents or multi-step workflows with a framework such as Google ADK, LangGraph, CrewAI, or the OpenAI Agents SDK.
• Awareness of agent tooling and protocols, including tool/function calling and the Model Context Protocol (MCP); familiarity with the A2A protocol is a plus.
• Basic understanding of agent harness concepts — session/state management, memory, and guardrails.
• Experience consuming major GenAI APIs (e.g., OpenAI, Gemini, Claude) and exposure to orchestration frameworks such as LangChain or LlamaIndex.
• Understanding of application deployment and containerization (Docker).
• Working knowledge of version control systems (Git).
• Awareness of AI compliance, data privacy, guardrails, and responsible AI principles.

 

Required Soft Skills

• Strong teamwork and communication abilities.
• Eagerness to learn new AI/GenAI and agentic technologies and frameworks.
• Analytical mindset and attention to detail.
• Openness to feedback, coaching, and continuous improvement.

 

Qualifications

• Bachelor's or Master's degree in Computer Science, Data Science, AI, or a related field.
• 3–6 years of professional experience in software/AI development, with exposure to Generative AI and agentic AI.
• Experience contributing to AI/GenAI or software projects in a collaborative team setting.
• Exposure to cloud-based AI/ML environments (AWS, GCP, or Azure) is a plus.

 

------------------------------------------------------

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.