Senior AI Engineer/AI Lead

We are seeking a Senior AI Engineer/AI Lead to join our team. As a Senior AI Engineer/AI Lead, you will drive the design and implementation of a cutting-edge, multi-agent AI platform that transforms how adverse event data is processed and managed in pharmacovigilance. This position plays a critical role in advancing clinical research automation, ensuring regulatory compliance, and establishing new standards for safety and efficiency in drug safety operations. You’ll collaborate with cross-functional teams including domain experts, quality assurance, and regulatory specialists, following industry-standard processes and contributing to system development from architecture through validation and deployment. Responsibilities include designing and implementing advanced AI agent orchestration, defining evaluation frameworks, architecting secure integrations with enterprise systems, and ensuring GxP compliance, with significant opportunities to influence technical direction, drive innovation in AI governance, and expand your expertise in regulated AI systems.

You’ll join a fast-paced, growth-oriented environment focused on making a meaningful impact through advancing pharmacovigilance automation, improving data accuracy and safety oversight, and establishing Parexel as a leader in AI-driven clinical research. With diverse teams and continuous learning opportunities, the Senior AI Engineer/AI Lead can explore emerging technologies, mentor junior engineers, and expand skills across AI architecture, regulatory compliance, and healthcare innovation.

Key Responsibilities

• Design and implement the multi-agent architecture using AWS Bedrock, AgentCore, Strands SDK, and/or LangGraph, with Anthropic Claude as the foundation model layer
• Define the agent topology including supervisor orchestration, inter-agent communication, state management, escalation routing, and comprehensive audit trail infrastructure
• Own the prompt engineering strategy across all pharmacovigilance agents, including system prompts, few-shot examples, and guardrails to ensure accuracy and compliance
• Architect, design, and build Model Context Protocol (MCP) servers to expose enterprise applications, data sources, and services as standardized tools for AI agents
• Build and maintain the evaluation pipeline by designing benchmarks, curating ground-truth datasets with domain experts, and running accuracy and precision measurements
• Architect the quality control layer including cross-model verification, deterministic rule engines for field validation, and auto-escalation logic
• Collaborate with the existing team to migrate current systems to Claude on Bedrock while preserving proven logic and re-engineering prompts and evaluation pipelines
• Define the CI/CD and MLOps strategy including model versioning, prompt version control, deployment of pipelines, monitoring dashboards, and cost tracking
• Own the technical validation strategy aligned to GAMP5 Category 5 requirements, working with QA to produce IQ/OQ/PQ documentation specific to LLM-based systems
• Write and review technical documentation including architecture decision records, algorithm descriptions, and AI model specifications aligned to regulatory frameworks

You’ll thrive in this role if you bring:

• Expertise in AI system architecture, prompt engineering, and large language model orchestration
• Experience designing and implementing secure integrations between AI systems and enterprise applications
• The ability to translate complex technical concepts into clear communication for non-technical stakeholders, including regulatory and quality audiences
• A commitment to quality, compliance, and patient safety in regulated environments
• Comfort working with AWS cloud services, Python, and modern AI frameworks and tools

Required Qualifications

• 7+ years of hands-on experience building ML/AI production systems, with at least 2 years working with large language models in application-level contexts
• Deep working knowledge of Anthropic Claude APIs, prompt engineering patterns, and retrieval-augmented generation architectures
• Practical experience with AWS, specifically Bedrock (model invocation, agents, knowledge bases), Lambda, S3, IAM, and CloudTrail
• Experience building multi-agent or multi-step LLM orchestration systems using frameworks such as LangGraph, LangChain, CrewAI, or Strands SDK
• Strong Python and software engineering fundamentals including API design, containerizatio…