Production AI Agents, Enterprise RAG, Python, and Azure/AWS
Mandatory Screening Criteria :Top PriorityCandidates must meet all of the following requirements:
- 10+ years of overall software engineering experience, with strong hands-on coding and system-design capability.
- Minimum 7 years of AI/ML experience, including recent, practical experience in Generative AI, LLMs, RAG, and Agentic AI solutions.
- 10+ years of product development experience, with ownership across design, development, testing, deployment, production support, and enhancement.
- Proven experience working as an individual contributor-able to independently design, develop, test, deploy, and troubleshoot production-grade AI applications.
- Prior experience in banking, financial services, fintech, payments, or other regulated financial environments is mandatory.
- Immediate joiners or candidates with very short notice periods are preferred.
- Hands-on experience using GitHub Copilot as part of the software-development lifecycle, including code development, test generation, code review support, documentation, and productivity improvement.
- Hands-on experience integrating Agentic AI solutions with enterprise tools and APIs, including Splunk, Grafana, bank internal applications, databases, document repositories, and workflow platforms.
- Experience building tool-enabled agents using secure API/function calling, authentication, access controls, audit logs, error handling, and human approval workflows.
- Strong expertise in Fast RAG and Graph RAG architectures, including hybrid retrieval, reranking, metadata filtering, knowledge graphs, semantic search, and context-grounded responses.
- Practical experience in LLM token optimisation, including prompt/context optimisation, intelligent chunking, retrieval filtering, model selection, caching, response streaming, latency optimisation, and cost monitoring.
- Experience with LLM harness engineering—designing reusable prompt templates, structured-output schemas, evaluation datasets, automated test harnesses, regression testing, and quality metrics for LLM/agent workflows.
- Knowledge of Model Context Protocol (MCP) for securely connecting AI agents to enterprise tools, data sources, and APIs through governed, reusable tool interfaces.
Role Overview
We are looking for a hands-on Senior Agentic AI Engineer to design, build, and deploy enterprise AI solutions that go beyond chatbots and prompt-based applications. The ideal candidate will have proven experience developing stateful, tool-enabled agents, supported by strong Python engineering, retrieval-augmented generation (RAG), evaluation, and cloud deployment skills.This is an engineering-focused role requiring ownership from architecture and implementation through testing, production deployment, and continuous improvement.
Key Responsibilities
- Design and implement multi-step agent workflows involving planning, tool selection, API execution, decision-making, and response validation.
- Build agent orchestration using LangGraph or equivalent frameworks, including state management, conversational memory, conditional routing, retries, and human-in-the-loop approvals.
- Integrate agents with enterprise REST APIs, databases, document repositories, and business applications using secure tool/function calling.
- Develop enterprise RAG pipelines covering document ingestion, semantic chunking, embeddings, hybrid retrieval, metadata filtering, reranking, and grounded responses with citations.
- Implement evaluation frameworks for agent task completion, tool-call correctness, retrieval quality, faithfulness, hallucinations, latency, and cost.
- Develop production-ready Python services using FastAPI, Docker, automated testing, CI/CD, logging, and monitoring.
- Apply security controls for prompt injection, sensitive-data handling, access permissions, and safe execution of agent actions.
- Collaborate with business stakeholders to translate requirements into reliable AI workflows with measurable outcomes.Role OverviewWe are seeking a hands-on Senior Agentic AI Engineer / AI Developer to build secure, scalable, production-grade AI solutions for banking and financial-services use cases.
- The role requires an experienced individual contributor who can independently translate business requirements into reliable Agentic AI products.
- The successful candidate must have delivered solutions beyond basic chatbots or prompt-based applications, including multi-step AI agents, tool/API calling, state management, RAG, evaluation frameworks, guardrails, and enterprise deployment.
- Key Responsibilities:
- Independently design, develop, test, deploy, and maintain Agentic AI applications for banking and financial-services use cases.
- Build multi-step AI agents that can plan tasks, call approved APIs/tools, use structured outputs, retain workflow state, manage failures, and escalate actions for human approval.
- Develop agent orchestration workflows using LangGraph, LangChain, or equivalent frameworks, including memory, conditional routing, retries, error handling, and auditability.
- Build enterprise-grade RAG platforms for policies, procedures, product documents, regulatory content, operational manuals, and knowledge repositories.
- Implement document ingestion, OCR/content extraction, chunking, embeddings, vector search, hybrid retrieval, reranking, metadata filtering, and response citations.
- Integrate LLM solutions securely with enterprise REST APIs, databases, workflow systems, document stores, and business applications.
- Develop Python-based backend services and APIs using FastAPI or equivalent frameworks.
- Use GitHub Copilot effectively and responsibly to improve engineering productivity, code quality, documentation, test coverage, and code-review efficiency.
- Implement LLM and agent evaluation frameworks covering task completion, answer relevance, retrieval quality, faithfulness, hallucination detection, tool-call accuracy, latency, and cost.
- Apply AI security and governance controls, including prompt-injection protection, PII/data masking, access control, permission-aware tool calling, audit logging, and human-in-the-loop approvals.
- Deploy containerised AI services using Docker, CI/CD pipelines, logging, monitoring, alerting, and performance optimisation.
- Work closely with product owners, architects, security teams, risk/compliance teams, and business stakeholders to deliver measurable outcomes.
- AI, Generative AI, and Agentic AI
- Minimum 7 years of practical AI/ML experience, with substantial recent experience in Generative AI and LLM application development.
- Demonstrable experience building production-grade Agentic AI solutions, not only chatbots or RAG-based Q&A applications.
- Hands-on experience with:
- LLMs, prompt engineering, structured outputs, and function/tool calling
- LangGraph, LangChain, or equivalent agent orchestration frameworks
- Agent state, conversational memory, workflow persistence, conditional routing, retries, and failure handling
- Human-in-the-loop approvals and escalation workflows
- Single-agent and multi-agent architecture patterns
- RAG, vector databases, semantic search, hybrid search, metadata filtering, and reranking
- LLM evaluation, LLM-as-a-Judge, hallucination detection, faithfulness testing, and human evaluation
Cloud, Deployment, and MLOps
- Experience with Azure OpenAI, OpenAI APIs, AWS Bedrock, or equivalent enterprise LLM platforms.
- Strong experience with Azure or AWS cloud services, preferably including secure enterprise deployments.
- Experience developing REST APIs using FastAPI, Flask, or equivalent frameworks.
- Experience with Docker, containerisation, CI/CD, environment management, logging, monitoring, and production support.
- Experience with vector databases such as Azure Cosmos DB Vector Search, Azure AI Search, Pinecone, Weaviate, FAISS, Elasticsearch, or equivalent.
- Experience with Databricks, MLflow, Unity Catalog, model governance, or comparable MLOps platforms is preferred.
Banking and Financial Services Requirements
- Mandatory experience delivering technology, AI, analytics, automation, or data solutions for banking, financial services, fintech, payments, insurance, lending, treasury, fraud, risk, compliance, or regulatory domains.
- Understanding of regulated-environment requirements, including data confidentiality, PII protection, audit trails, access controls, governance, model risk, and secure API integration.
- Experience with banking use cases such as customer service, operations automation, compliance, KYC/AML, fraud detection, risk assessment, document intelligence, financial-product knowledge, or internal employee assistants is highly desirable.
Candidate Submission RequirementsCandidates should provide a brief description of at least one production Agentic AI solution they personally developed, including:
- Architecture: Agent workflow, orchestration framework, tools/APIs, RAG components, state/memory design, and individual contribution.
- Reliability and security: Guardrails, permissions, failure handling, human approvals, logging, evaluation approach, and compliance controls.
- Production delivery: Cloud platform, deployment approach, CI/CD, user scale, and measurable business, quality, latency, or cost outcomes.
- Banking/financial experience: Relevant client, project type, use case, and responsibilities, without disclosing confidential information.
Pay: ₹2,000,000.00 - ₹3,500,000.00 per year
Benefits:
- Health insurance
- Paid sick time
- Provident Fund
Experience:
- artificial intelligence: 7 years (Required)
- Software development: 10 years (Required)
- agentic AI: 7 years (Required)
- Product development: 10 years (Required)
Location:
- Hyderabad, Telangana 500050 (Required)
Shift availability:
- Day Shift (Required)
- Overnight Shift (Required)
Work Location: In person
