Lead AI/ML Consultant

We are looking for a Lead AI/ML Consultant (Agentic AI Engineer) for the ADP account engagement. This is a pivotal technical position centered on architecting and building agentic AI solutions powered by AWS services. The resource will head up the AI/ML workstream, partnering with Technical Business Analysts to bring GenAI capabilities to life for the client.

Responsibilities

• Architect and build multi-agent AI systems, encompassing orchestration patterns, agent-to-agent delegation, tool/function calling, and memory management
• Develop agentic AI solutions leveraging Amazon Bedrock Agents, Knowledge Bases, Guardrails, and Flows
• Deploy solutions using agentic frameworks including LangGraph, LangChain, CrewAI, AutoGen, or comparable tools
• Architect and build Retrieval-Augmented Generation (RAG) systems utilizing vector stores such as OpenSearch, Kendra, and Pinecone
• Create and refine prompts for diverse LLM use cases, including evaluation workflows and hallucination reduction techniques
• Integrate with various LLM providers, such as Claude, Amazon Nova, and Llama, through the Amazon Bedrock platform
• Connect AI solutions with serverless components such as Lambda, Step Functions, API Gateway, S3, and DynamoDB
• Establish guardrails, PII handling protocols, and enterprise compliance measures to ensure secure and responsible AI operations
• Architect token-efficient designs and apply cost governance strategies to manage and optimize AI solution spending

Requirements

• A minimum of 5 years of relevant experience
• At least one year of experience leading and managing teams
• Background in building agentic and multi-agent workflows that support orchestration, delegation, and autonomous task execution among AI agents
• Direct experience with Amazon Bedrock for developing and deploying generative AI solutions, including agents, knowledge bases, and guardrails
• Applied experience with LangChain/LangGraph for constructing and coordinating LLM-powered applications and agentic workflows
• Experience architecting RAG (Retrieval-Augmented Generation) systems to anchor LLM outputs in accurate, relevant data
• Advanced Python skills for building AI/ML applications, integrating models, and developing automation scripts
• Experience in prompt engineering, covering the design, testing, and refinement of prompts to enhance LLM performance and reliability
• Excellent English communication skills (B2 level or higher)

Nice to have

• Experience with Amazon SageMaker for developing, training, and deploying machine learning models
• Exposure to LLMOps practices for overseeing the full lifecycle of LLM-based applications, from deployment through monitoring and ongoing improvement
• Understanding of vector databases for storing and retrieving embeddings to enable RAG and semantic search functionality