Key Responsibilities:

- Lead a team of GenAI engineers — providing technical leadership, architecture guidance, mentoring, and code reviews to drive engineering excellence and delivery quality.

- Actively contribute production -quality code — this is a hands -on role; you will design, develop, and deploy GenAI solutions alongside the team, not just guide from the sidelines.

- Design and deploy production -grade GenAI and Agentic AI applications — secure, scalable, compliant, and aligned with BFSI regulatory requirements (RBI, SEBI, GDPR).

- Build AI agents and multi -agent workflows using Amazon Bedrock, Bedrock Agents, AgentCore, and modern agent orchestration frameworks — enabling autonomous, multi -step financial AI processes.

- Design and implement RAG -based solutions — including vector database integration, chunking strategies, retrieval optimisation, prompt engineering, AI guardrails, and model evaluation frameworks.

- Develop reusable frameworks, libraries, and accelerators — standardising GenAI engineering practices and improving delivery velocity across the team.

- Design and implement cloud -native solutions on AWS — using services such as Bedrock,



SageMaker, Lambda, S3, and Infrastructure as Code (Terraform) for scalable, production -ready deployments.

- Collaborate with Solution Architects, Product Owners, and business stakeholders — translating complex BFSI requirements (fraud detection, risk analytics, compliance, customer engagement) into scalable GenAI solutions.

- Troubleshoot complex technical issues — from hallucination rates and retrieval quality to latency, cost optimisation, and production incidents — guiding teams through design and implementation challenges.

- Produce high -quality technical documentation — including architecture documents, HLDs, LLDs, API documentation, deployment guides, and operational runbooks for audit and knowledge sharing.

- Stay current with emerging AI technologies — evaluating current models, frameworks, and approaches; recommending improvements to architecture, engineering practices, and solution design.


Requirements
- Amazon Bedrock expertise: deep hands -on experience with Bedrock, Bedrock Agents, AgentCore, Knowledge Bases, and Bedrock Guardrails for production GenAI applications.

- Agentic AI development: proven experience building multi -agent systems and autonomous workflows using LangChain, LangGraph, LlamaIndex, CrewAI, AutoGen, or AWS -native agent frameworks.

- RAG pipeline design: end -to -end RAG implementation — vector databases (Pinecone, OpenSearch, Weaviate, pgvector), embedding models, retrieval strategies, and evaluation using RAGAS or equivalent.

- Prompt engineering mastery: advanced prompting techniques — few -shot, chain -of -thought, structured outputs, system prompt design,



and prompt versioning for production systems.

- Python proficiency: production -grade Python for GenAI applications, API development (FastAPI), and integration with AWS services using Boto3.

- AWS cloud architecture: practical experience with AWS services — Bedrock, SageMaker, Lambda, S3, DynamoDB, API Gateway, ECS/EKS — and Terraform for IaC.

- LLM evaluation and observability: model evaluation frameworks, A/B testing, hallucination monitoring, latency tracking, cost optimisation, and LLMOps practices in production.

- Technical leadership: proven track record leading engineering teams — architecture reviews, code reviews, mentoring, and driving delivery standards in an Agile environment.

- BFSI domain understanding: familiarity with banking and financial services use cases — fraud detection, AML, credit risk, regulatory reporting, customer analytics, and compliance constraints (RBI, SEBI, GDPR).​