Senior AI Engineer – Agentic AI & RAG

Job description

**Band:** U4 / P1 **Experience:** 10–12 years **Location:** Remote (India) **Employment Type:** Full-Time **Role Overview** We are looking for a **Senior GenAI Engineer with strong hands-on experience in building production-grade Generative AI solutions** . The role requires deep expertise in **agentic AI architectures, LLM orchestration, and advanced RAG pipelines** , with proven experience in **deploying scalable, enterprise-grade solutions (not PoCs).** **Key Responsibilities** - Design and build **end-to-end production-grade GenAI applications** using LLMs - Develop and orchestrate **agentic AI systems (single agent & multi-agent)** for complex enterprise workflows - Implement **RAG pipelines** including document ingestion, embeddings, retrieval optimization, and response synthesis - Build and optimize **LLM orchestration workflows** with strong focus on latency, cost, and scalability - Implement **observability frameworks** (tracing, monitoring, logging) for GenAI systems - Define and execute **evaluation frameworks** for LLM response quality, grounding, and hallucination management - Develop scalable backend services using **Python + FastAPI** - Build lightweight UI layers using **Streamlit** for demos/internal tools - Ensure **production readiness** including scalability, resilience, and fault tolerance - Collaborate with architecture, data, and platform teams to integrate GenAI into enterprise ecosystems **Mandatory Skills (Strict – No Compromise) – REJECT PROFILES IF ANY OF THE BELOW MENTIONED IS MISSING** - Proven experience in **building production-grade GenAI solutions** (must demonstrate real deployments) - Hands-on expertise in **agentic frameworks & orchestration** : - AWS AgentCore / Strand Agents - LangGraph (or equivalent agent orchestration frameworks) - Multi-agent system design - Strong hands-on experience with **RAG architectures** : - Vector DBs (FAISS, Pinecone, OpenSearch, Chroma, etc.) - Embeddings & retrieval strategies (hybrid search, reranking, grounding) - Deep understanding of **LLM orchestration workflows** - Experience in **LLMOps / Observability / Evaluation** : - Monitoring LLM performance, tracing, logging - Evaluation frameworks (RAGAS, DeepEval or equivalent) - Strong coding expertise in **Python** - Experience building APIs using **FastAPI** **Good-to-Have** - Experience with AWS Bedrock / SageMaker-based GenAI deployments - Exposure to guardrails, prompt injection handling, and GenAI risk controls - Knowledge of **cost optimization & token efficiency strategies** - Experience in enterprise domains (BFSI, Pharma, Healthcare, Insurance) - CI/CD and containerized deployment (Docker/Kubernetes) **Profile Expectations (Important for Screening)** - Must clearly explain **architecture of at least 1–2 production GenAI implementations** - Should demonstrate **ownership of solution design (not just usage of APIs/frameworks)** - Strong depth in **agent workflows (planner-executor, tool-calling, multi-agent orchestration)** - RAG understanding should go beyond chatbot-level (must include retrieval tuning & grounding logic)