AI Agent Developer

(Please be informed that this is a Contract role. The role can be based either in CHENNAI or BANGALORE.)

ABOUT EMIRATES NBD:

Emirates NBD is a market leader across the MENAT (Middle East, North Africa and Türkiye) region with a presence in 13 countries, serving over 9 million customers. Emirates NBD is the leading financial services brand in the Emirates with a Brand value of approximately USD 4 billion.

We serve our customers and help them realize their financial objectives through a range of banking products and services including retail banking, corporate & institutional banking, Islamic banking, investment banking, private banking, asset management, global markets and treasury, and brokerage operations.

We are a key participant in the global digital banking industry, with 97% of all financial transactions and requests conducted outside of our branches. We also operate Liv, the lifestyle digital bank by Emirates NBD. With over half a million users, it continues to be the fastest-growing digital bank in the region.

JOB PURPOSE

The AI Agent Developer independently translates business requirements into working AI agent solutions from prompt design to tool/skill creation, API integrations, and workflow automation.

The ideal candidate is hands-on, self-driven, and comfortable owning a problem from requirement gathering through deployment.

KEY RESPONSIBILITIES

• Work directly with business stakeholders to understand requirements and convert them into functional agent solutions on Oracle Fusion HCM Cloud using Oracle AI Agent Studio
• Design, write, and refine prompts (system prompts, few-shot examples, chain-of-thought structuring) for reliable, consistent agent behavior
• Build and maintain custom tools/skills/functions that agents can call (e.g., data lookups, calculations, document generation, third-party actions)
• Integrate agents with external and internal APIs (REST/SOAP) - authentication, request/response handling, error handling, retries
• Design and implement multi-step workflows/orchestration (sequential, conditional, parallel agent/tool calls)
• Set up and manage workflow triggers — event-based, schedule-based, webhook-based, or user-initiated
• Test agent behavior thoroughly: edge cases, failure handling, hallucination checks, guardrails
• Optimize for cost, latency, and accuracy (model selection, context management, caching)
• Document agent architecture, prompts, tools, and workflows for maintainability
• Debug and troubleshoot agent/tool/workflow failures in production
• Stay current with LLM/agent framework updates and recommend improvements.

KEY QUALIFICATIONS

Skills & Experience

• 3–5 years’ experience building LLM-based applications or AI agents (production experience preferred)
• Strong prompt engineering skills — system prompts, role-based prompting, few-shot/zero-shot design, output formatting (JSON/XML), guardrails against prompt injection
• Hands-on experience creating custom tools/functions/skills for agent frameworks (function calling / tool use)
• Practical API experience — calling REST/SOAP APIs, handling auth (OAuth, API keys), parsing responses, error handling
• Experience setting up triggers — webhooks, cron/schedule-based, event-driven triggers
• Understanding of vector databases/RAG (retrieval-augmented generation) is a plus
• Ability to independently scope a requirement, propose a solution approach, and execute without heavy supervision

Preferred Qualifications

• Experience building multi-agent systems (agent-to-agent handoff, supervisor/worker patterns)
• Exposure to MCP (Model Context Protocol) or similar agent-tool standards
• Experience with monitoring/observability for AI agents (logging, tracing, evaluation frameworks)
• Familiarity with version control (Git) and CI/CD for deploying agent updates
• Node.js & TypeScript
• Understanding of Microservice architecture. Practical experience in developing microservices and maintaining on production.
• Clear understanding of secure identity protocols like SSO and OAuth.
• Solid understanding of Relational and Non-relational database operations.
• Understanding of core LLM Fundamentals concepts like tokens, top_p, temperature, and streaming data.
• Hands-on experience implementing structured outputs, tool callings, and function callings.
• Ability to articulate solutions with multiturn tool calls aka agentic solution.
• Production experience building workflows using LangChain and/or LangGraph.
• Explain the architectural pattern Retrieval-Augmented Generation (RAG) and advanced prompting techniques.
• Experience on Oracle AI Agent Studio will be an added advantage

Soft Skills

• Strong ownership mindset — can take a vague business ask and turn it into a working solution
• Good written communication for documenting logic and prompts clearly
• Analytical thinking to debug non-deterministic (LLM) behavior systematically
• Comfortable iterating quickly based on testing/feedback

There’s never been a better time to join Emirates NBD.

We’re one of the region’s most recognizable brands. We’re banking innovation leaders. We’re growing across both the UAE and our global offices. We offer a huge range of professional development opportunities to accelerate your career.

It also goes without saying that we provide extremely competitive rewards, benefits and perks too, like our flexible work policy so you can work from home whenever it suits.

At ENBD, we encourage interested candidates to review the key responsibilities and qualifications for each role and apply for positions that match their skills and capabilities.