Principal AI Engineer

Overview

Lead the modernization of Johnson & Johnson's enterprise Generative AI platform on Amazon Web Services, moving GenAI capabilities off bespoke infrastructure onto AWS cloud-native services, and engineer the AWS foundation the platform runs on. Based in Raritan, NJ, reporting to the Senior Manager, TS GenAI platform engineering.

What you'll do

• Own the GenAI cloud-native modernization roadmap on AWS: assess each service against its managed equivalent, decide replace, refactor, retain or retire, and re-platform onto Bedrock, AgentCore, ECS and EKS, Lambda, Step Functions, DynamoDB and S3.
• Define the AWS Well-Architected reference architecture and enforce a cloud-native-by-default standard, favoring managed orchestration over bespoke agent development.
• Engineer the AWS foundation: Bedrock access and quotas; the landing zone and multi-account architecture on Organizations and Control Tower; VPC, PrivateLink, Transit Gateway and least-privilege IAM, delivered for standardized consumption with the Cloud and ISRM teams.
• Codify the estate as Infrastructure as Code in Terraform and CDK, including modules, pipelines, drift detection and policy-as-code, replacing console changes and contributing hands-on where needed.
• Run migration waves without disrupting users: Route 53 weighted and blue-green cutover, backward-compatible APIs, data migration, documented rollback, and decommissioning only once the AWS replacement is proven in production.
• Deliver GxP-qualified AWS infrastructure under change control agreed with Quality and ISRM, and partner with GenAI governance on guardrail, identity and traceability controls.
• Engineer the hosting substrate for enterprise platform capabilities (APIM, Content Sphere, Agent Framework, Drafting Assist, the MCP Gateway and the Agent, Skills and MCP catalog) plus the semantic layer, retrieval and evaluation harness services.
• Own cost, performance and resilience engineering: tagging and showback, commitment discounts, Bedrock inference efficiency, multi-AZ and multi-region design, capacity planning and disaster recovery, with CloudWatch and X-Ray observability and reporting.
• Lead, coach and develop the AWS platform engineering team in an inclusive, Credo-grounded environment, accountable for timely delivery, budget optimization and engineering quality, and supporting LFRP planning.
• Own the AWS partnership and cross-cloud alignment: architecture reviews, early access and quota negotiation, position papers on emerging capabilities, adoption of agentic approaches in software development, and parity with the Azure platform engineering sub-team on shared primitives.

What you'll need

• Minimum of 8 years of progressive experience in cloud infrastructure or platform engineering, including at least 2 years of formal people-leadership experience managing engineers, or demonstrated equivalent leadership of engineering teams.
• Bachelor's degree in Computer Science, Engineering, Information Technology or a related technical discipline is required.
• Demonstrated hands-on ownership of a production AWS estate at enterprise scale, including multi-account architecture, landing zone design and the operational reality of running it.
• Proven experience delivering infrastructure through Infrastructure as Code as the primary interface, with a track record of eliminating manual console changes in favor of reviewed, versioned code.
• Experience engineering cloud infrastructure in a regulated environment (GxP, validated systems, or an equivalent regime such as SOX or a comparable audited control framework) and producing evidence that satisfies auditors.
• Practical experience supporting AI, machine learning or Generative AI workloads on AWS, including the networking, identity and capacity considerations specific to managed inference services.
• Demonstrated ability to manage cloud cost and performance as engineering problems, with specific examples of architectural changes that improved efficiency without degrading service.
• Experience working directly with a hyperscale cloud provider's account and solutions-architecture teams to influence quotas, service adoption and roadmap alignment.
• Ability to communicate infrastructure trade-offs clearly to both engineers and non-technical stakeholders, including Quality, Security and business partners.
• Experience coaching senior individual contributors and supporting their technical growth without taking over their technical decisions.
• Required skills: cloud-native modernization and re-platforming; AWS Organizations, Control Tower, service control policies and multi-account landing zone design for regulated workloads; Amazon Bedrock platform configuration (model access, inference profiles, provisioned throughput, service quotas and private endpoints); AWS compute and data services (ECS and EKS, Lambda, Step Functions, DynamoDB, S3, API Gateway); semantic layer and governed data access on AWS (Bedrock Knowledge Bases, Athena or Glue catalog integration); AWS networking (VPC and subnet architecture, PrivateLink and VPC endpoints, Transit Gateway, Route 53, egress control); AWS IAM in depth (role and trust policy design, permission boundaries, least privilege, cross-account access, identity federation); Infrastructure as Code with Terraform and AWS CDK (module design, state management, policy gates, CI/CD delivery); migration engineering (dual-run and cutover planning, backward-compatible interfaces, data and state migration, rollback and decommissioning); AWS cost engineering; AWS reliability engineering (multi-AZ and multi-region design, capacity planning, disaster recovery); GxP qualification of cloud infrastructure and IaC change; contractor and vendor management.

Nice to have

• Experience leading a large-scale re-platform or cloud migration of an established production service, including legacy decommissioning.
• Experience running Microsoft Azure alongside AWS and a practical view of where cross-cloud consistency is worth its cost.
• AWS certifications such as Solutions Architect Professional, DevOps Engineer Professional or Machine Learning Specialty.
• Kubernetes platform engineering with GitOps delivery, cluster lifecycle management and multi-tenancy.
• Familiarity with Amazon Bedrock AgentCore sufficient to support agentic workloads on the platform.
• Computer system validation for cloud infrastructure in a pharmaceutical or medical device setting.
• Familiarity with Model Context Protocol and the AWS infrastructure requirements of hosting MCP servers securely.
• Experience with FinOps practice for AI and inference workloads.
• Preferred: a Master's degree or a Ph.D.

Details

• Location: Raritan, New Jersey, United States of America.
• Anticipated base pay range: $102,000.00 - $177,100.00.