AWS Gen AI Manager

Key Responsibilities

• Lead the design and architecture of generative AI applications using Amazon Bedrock, Bedrock AgentCore, and SageMaker.
• Architect production-grade RAG pipelines using Bedrock Knowledge Bases with agentic retrieval, hybrid search, reranking, and multi-modal content (text, images, audio, video).
• Drive custom ML model development and fine-tuning strategies using SageMaker HyperPod, JumpStart, and Unified Studio.
• Define and implement prompt engineering frameworks; optimize cost with Intelligent Prompt Routing and Model Distillation (up to 500% faster, 75% cost reduction).
• Architect multi-agent orchestration patterns using Bedrock AgentCore with supervisor agents, RAG integration, code interpretation, and memory retention.
• Establish Bedrock Guardrails policies for responsible AI including content filtering, PII protection, Automated Reasoning checks, and hallucination mitigation.
• Lead the design of scalable data pipelines for AI/ML workloads using SageMaker Data Processing and Bedrock Data Automation.
• Establish AI/ML best practices, governance frameworks, and responsible AI guidelines.
• Lead and mentor GenAI engineering teams, driving technical excellence.
• Collaborate with product, data science, and platform teams to deliver AI-driven solutions.
• Evaluate emerging GenAI technologies (new foundation models, MCP integrations) and recommend adoption strategies.

Required Qualifications

• Bachelor's or Master's degree in Computer Science, Data Science, AI/ML, or related field.
• 10+ years of experience in software engineering or ML engineering.
• 5+ years of hands-on experience with AWS AI/ML services.
• Deep expertise in GenAI application architecture using Bedrock, Bedrock AgentCore, and foundation models.
• Extensive experience with SageMaker for custom model development, fine-tuning, MLOps, and HyperPod distributed training.
• Proven experience architecting production RAG pipelines with Bedrock Knowledge Bases (agentic retrieval, multi-modal support) and enterprise-grade reliability.
• Expert-level Python and ML frameworks (PyTorch, TensorFlow, Hugging Face).
• Strong experience with orchestration frameworks (LangChain, LlamaIndex, CrewAI, Strands Agents) and multi-agent patterns.
• Strong experience in prompt engineering, Model Distillation, Intelligent Prompt Routing, and evaluation at scale.
• Deep understanding of LLM architectures, fine-tuning approaches, RLHF, and inference optimization.
• Experience with Bedrock Guardrails, Automated Reasoning checks, and responsible AI practices.
• Excellent communication, leadership, and stakeholder management skills.
• Knowledge of Agile Methodology and Practices.