Role Overview

Max Healthcare is seeking an experienced AI Engineer with a strong background in Large Language Models (LLMs) and Retrieval-Augmented Generation (RAG). This role is crucial for architecting, developing, and deploying advanced AI agents and RAG pipelines. You will be responsible for integrating vectorized knowledge bases and building scalable, AI-powered applications using Python, React, and cloud-native AWS services.

Responsibilities

• Design, build, and optimize RAG workflows incorporating vector databases and embeddings.
• Fine-tune, prompt-engineer, and deploy various LLMs (e.g., Amazon Bedrock, GPT, Llama, Claude) for specific production use cases.
• Develop robust Python backend services, including APIs and orchestration layers, and create intuitive React user interfaces for AI applications.
• Deploy and manage AI workloads on AWS infrastructure, utilizing services like EKS/ECS, Aurora PostgreSQL, OpenSearch with pgvector, SQS, Lambda, Elastic Cache, and Secrets Manager.
• Implement essential guardrails, monitoring systems, and evaluation frameworks to ensure the safe, reliable, and compliant operation of AI agents.
• Collaborate closely with product management and infrastructure teams to guarantee the scalability, performance, and regulatory compliance of AI solutions.

Skills & Experience

• A minimum of 2-4 years of experience in AI/ML engineering, with demonstrated project success in LLM and RAG implementation.
• Proficiency in Python for developing AI pipelines and APIs, with exposure to React.js.
• Hands-on experience with AWS AI and infrastructure services, including EKS/ECS, Aurora PostgreSQL, OpenSearch/pgvector, SQS, Lambda, S3, and Secrets Manager.
• Solid understanding of database technologies such as PostgreSQL, MongoDB, Neo4j, and Vector Databases.
• Familiarity with popular AI frameworks like Lang-Chain, Llama-Index, and Hugging Face.
• Good knowledge of containerization technologies (Docker, Kubernetes) and CI/CD practices (CDK, Code-Pipeline).
• Understanding of asynchronous processing, Dead Letter Queue (DLQ) handling, and scalable message-driven architectures.

Must-Have Qualifications

• Proven experience with Amazon Bedrock and building LLM-based agents.
• Exposure to MLOps practices and tools (e.g., MLflow, Weights & Biases).
• Experience with hybrid architectures, combining graph databases (Neo4j) and vector databases for RAG.
• Demonstrated experience with prompt guardrails, structured output validation, and secure data handling protocols.