AI Engineer

Senior AI Engineer Role - 7+ Years Experience Position Overview: We are seeking an experienced AI Engineer to design, build, and deploy production-grade AI and LLM applications on AWS. This senior role demands deep hands-on expertise in retrieval-augmented generation (RAG), AI agents, backend development, and secure cloud-native deployment. You will bridge the gap between AI prototypes and scalable, reliable, observable production systems — working closely with product, DevOps, and platform teams to deliver enterprise-grade intelligent applications. Key Responsibilities: - Own the full lifecycle of AI/LLM applications from architecture through deployment, monitoring, and continuous improvement - Design and optimize end-to-end RAG pipelines including document ingestion, chunking, embedding generation, vector storage, retrieval, reranking, and LLM synthesis - Build and orchestrate multi-agent AI systems using frameworks like LangChain, LangGraph, AutoGen, and CrewAI - Develop backend APIs and microservices using FastAPI with production-grade reliability and performance - Implement observability, tracing, and monitoring specifically designed for LLM and AI workloads - Design security guardrails, content filtering, hallucination mitigation, and responsible AI controls - Make independent technical decisions and mentor less experienced engineers - Drive best practices across the team for production AI systems Core AI & LLM Expertise Required: - Deep, practical production experience with foundation models (OpenAI GPT, Anthropic Claude, Meta Llama, Amazon Nova, Google Gemini) - Understanding of model selection, fine-tuning strategies, cost-latency trade-offs - Advanced prompt engineering including systematic prompt design, structured outputs, function/tool calling, and evaluation frameworks - Hands-on experience with streaming AI responses using SSE or WebSockets for real-time applications - Chain-of-thought, ReAct, and other advanced prompting patterns RAG & Vector Search Systems: - End-to-end RAG pipeline design and optimization experience - Proficiency with vector databases: Pinecone, ChromaDB, Qdrant, Weaviate - Knowledge of HNSW indexing, IVF-PQ quantization, similarity metrics (cosine, dot product, Euclidean) - Experience with hybrid search combining dense vectors with sparse BM25 retrieval - Amazon OpenSearch for search capabilities - Embedding model selection, chunking strategies, reranking architectures - Retrieval evaluation metrics: Recall@K, MRR, NDCG AI Agents & Orchestration: - Genuine production experience building AI agents with LangChain, LangGraph, AutoGen, CrewAI, or Semantic Kernel - Memory management, tool selection, error handling, and retry logic - Multi-agent systems design including communication protocols and coordination patterns - Model Context Protocol (MCP) servers and client integrations - Agent evaluation methodologies including success rates, task completion quality, and cost per interaction Backend Engineering & AWS Infrastructure: - Expert Python developer with deep FastAPI production experience - Async Python, connection pooling, performance profiling, OpenAPI schema design - Strong SQL and relational database design skills (PostgreSQL, MySQL, Aurora) - Extensive AWS experience: Bedrock, S3, Lambda, ECS/EKS, OpenSearch, RDS, DynamoDB - AWS networking, IAM roles and policies, VPC design, cost optimization - Git, GitHub, GitHub Actions for CI/CD pipelines, code review practices Observability & Evaluation: - LLM-specific observability: token counts, model selection, prompt versions, retrieval quality, agent decision traces - Experience with Langfuse, Arize Phoenix, Helicone, or LangSmith - Evaluation frameworks: Ragas, DeepEval, LangSmith evaluation capabilities - Implementation of systematic RAG and agent quality measurement - Responsible AI: guardrails, content filtering, PII detection, hallucination mitigation - Enterprise security, compliance requirements, governance frameworks Nice-to-Have Skills: - Multi-cloud experience (Azure OpenAI, Google Vertex AI) - Docker, Kubernetes, Terraform, CloudFormation - Node.js and TypeScript for full-stack AI applications - MLOps/LLMOps CI/CD pipelines, model versioning, automated testing - Enterprise security standards: SOC 2, ISO 27001, GDPR compliance - Multi-agent systems and autonomous workflow architecture Working Conditions: - Night shift schedule: 5:00 PM to 2:00 AM IST (intentional alignment with global collaboration windows) - On-site presence required at one of eight India locations - Collaborative, cross-functional teams with product, DevOps, and platform engineers - Focus on production-grade systems, not prototypes or proof-of-concepts Qualifications: - 7+ years in software engineering, AI, ML, or backend development - Demonstrated hands-on experience with production AI systems - Portfolio work, open-source contributions, and detailed case studies valued - Must confirm availability for night shift schedule before applying