3 days ago
Pune, IndiaStaff+
Responsibilities
- Design and implement generative AI and agentic AI solutions for enterprise business challenges.
- Architect context engineering strategies, including context layering, chaining, compression, pruning, offloading, and memory management.
- Build prompt-engineering, RAG, hybrid-search, multi-vector retrieval, re-ranking, knowledge graph, and Graph RAG systems.
- Design agentic workflows, multi-agent systems, agent harnesses, governance controls, feedback loops, state management, and execution controls.
- Integrate agents with tools and data using MCP and coordinate inter-agent collaboration using A2A.
- Support production deployment, scalability, observability, maintainability, and real-time or streaming AI solutions.
- Apply ethical AI guidelines, guardrails, agent isolation, sandboxing, data privacy, and compliance standards.
- Mentor junior team members, conduct code reviews, share knowledge, and collaborate with cross-functional stakeholders.
Requirements
- Bachelor's or master's degree in Computer Science, Data Science, AI, or a related field.
- 5–7 years of experience in AI or software development, including significant Generative AI and agentic AI experience.
- Deep hands-on knowledge of foundation models, LLMs, embeddings, tokenization, and context-window management.
- Advanced prompt and context engineering experience, including dynamic context orchestration.
- Experience building RAG systems, knowledge graphs, Graph RAG pipelines, and multi-hop retrieval using technologies such as Neo4j or ArangoDB.
- Experience with Google ADK or comparable agent frameworks such as LangGraph, Microsoft Agent Framework, CrewAI, or OpenAI Agents SDK.
- Experience with multi-agent orchestration, tool and function calling, planning, memory, harness engineering, agent isolation, and sandboxing.
- Hands-on experience with MCP and A2A protocols, agent observability, evaluation, and OpenTelemetry-based tooling.
- Proficiency with OpenAI, Gemini, and Claude APIs, as well as LangChain and LlamaIndex.
- Strong NLP skills, including NER, dependency parsing, text classification, and topic modeling.
- Proficiency with vector databases, embedding models, Python, data preprocessing, document ingestion, and API development.
- Experience with Docker, Kubernetes, CI/CD pipelines, and AWS or equivalent cloud infrastructure.
- Demonstrated portfolio of successful AI-driven business projects and the ability to collaborate, communicate, solve problems, learn, and mentor.
Tech Stack
Categories
About Citi
Citi's mission is to serve as a trusted partner to our clients by responsibly providing financial services that enable growth and economic progress. Our core activities are safeguarding assets, lending money, making payments and accessing the capital markets on behalf of our clients. We have over 200 years of experience helping our clients meet the world's toughest challenges and embrace its greatest opportunities. We are Citi, the global bank – an institution connecting millions of people across hundreds of countries and cities. For information on Citi’s commitment to privacy, visit on.citi/privacy.