3 days ago
Pune, IndiaStaff+
Responsibilities
- Translate complex business workflows and requirements into technical specifications and AI solution designs.
- Design, prototype, evaluate, and deploy multi-agent architectures, agentic tool orchestration, dynamic RAG retrieval engines, and human-in-the-loop workflows.
- Convert ambiguous business problems into functional proofs of concept within 2–4 week cycles and harden them for enterprise deployment.
- Develop custom scripts, integrations, and workflows for AI/GenAI applications.
- Integrate agent services with backend systems, user interfaces, DevOps infrastructure, transactional systems, and client lifecycle management platforms.
- Troubleshoot model and system integration issues and recommend tailored technical solutions.
- Gather field feedback, identify model-performance issues, and inform AI product roadmaps.
- Train business and technical users on prompt engineering, model interaction, and operational maintenance.
- Identify AI limitations, privacy, bias, ethical, compliance, and business-impact risks.
- Lead technical documentation and approval activities for Model Risk Management, Responsible AI, and InfoSec processes.
- Advise or coach junior analysts and operate independently as a subject-matter expert for senior stakeholders.
Requirements
- 7+ years of software engineering experience, including at least 3+ years in applied NLP, GenAI, LLM application development, and autonomous agent orchestration.
- Advanced Python proficiency with FastAPI, Asyncio, Pydantic, LangChain, and LangGraph.
- Strong working knowledge of Java with Spring Boot or modern TypeScript/Node environments.
- Experience building multi-agent systems, ReAct loops, Plan-and-Solve patterns, stateful workflows, memory management, and structured JSON output parsing.
- Hands-on experience with vector search engines including PostgreSQL with pgvector, Qdrant, Milvus, or Pinecone, as well as relational and NoSQL datastores.
- Familiarity with Claude, GPT-4, Gemini, R2D2, AWS Bedrock, and Azure OpenAI, including token optimization, rate limiting, and cost modeling.
- Experience with Docker, Kubernetes or OpenShift, CI/CD pipelines, AWS or GCP services, and enterprise observability using distributed tracing and OpenTelemetry.
- Ability to scope and deliver complex AI projects under ambiguity with minimal upfront requirements.
- Strong executive communication skills for explaining AI architecture trade-offs, confidence intervals, and ROI.
- Preferred experience includes Institutional Banking, FinTech, Wealth Management, KYC/AML, Fund Operations, Model Risk Governance, Fair Lending, or Regulatory AI Compliance.
- Working knowledge of Model Context Protocol or enterprise tool-calling frameworks is preferred.
- Open-source agentic AI contributions, research publications, or community AI toolkit experience are bonuses.
Tech Stack
AWSDockerFastAPIGoogle Cloud PlatformJavaKubernetesNode.jsOpenShiftPostgreSQLPythonSpring BootTypeScript
Categories
Forward Deployed
About Citi
Citi is a public financial-services company offering consumer and institutional banking, credit cards, wealth management, treasury and trade solutions, and capital-markets services. It serves individuals, corporations, financial institutions, and governments in more than 160 countries and jurisdictions, earning interest and fee income from lending, payments, trading, and advisory. Founded in 1812 and headquartered in New York, it trades on the NYSE under the ticker C.