Join EY GDS Managed Services' "AI for AI" team as an AI Platform Engineering Manager. This role is pivotal in deploying, scaling, and ensuring the reliability of our agentic AI solutions across a complex hybrid infrastructure. You will lead a team dedicated to transforming AI prototypes into robust, enterprise-grade platforms that operate globally, securely, and cost-effectively.

This position sits at the exciting intersection of platform engineering, MLOps/LLMOps, and agentic systems architecture, offering a unique opportunity to shape the future of AI infrastructure.

You will own the end-to-end platform deployment for agentic AI solutions across AWS, Azure, and GCP, as well as on-prem and hybrid environments. Key duties include designing and managing CI/CD and MLOps/LLMOps pipelines for seamless model deployment, versioning, and continuous evaluation.

Architecting for scale and resilience to ensure platforms reliably serve over 100,000 concurrent users globally is crucial. You will lead a team of engineers, providing technical direction and mentorship, while partnering closely with AI Engineers, Forward Deployed Engineers, Responsible AI leads, and Product Design to translate complex workflows into production-ready systems.

Manage the model-hosting strategy, considering trade-offs between hosted APIs and self-hosted models for data residency needs. Build comprehensive observability and monitoring for agentic systems, covering tracing, logging, cost, drift detection, and performance dashboards. Drive security, compliance, and governance in collaboration with InfoSec and Responsible AI teams, especially for regulated clients.

Optimize infrastructure cost and performance, balancing GPU and compute resource allocation. Contribute significantly to the platform strategy and roadmap, advising leadership on build-vs-buy decisions and future infrastructure investments.

We are seeking candidates with 5–8 years of experience in platform engineering, infrastructure/DevOps, SRE, or MLOps, including at least 2 years focusing on AI/ML or LLM-based systems in production. Proven experience deploying and scaling software/AI platforms across multiple cloud providers (AWS, Azure, GCP) and in on-prem/hybrid environments is essential.

Demonstrated experience deploying and managing both hosted LLM APIs and self-hosted/open-source models in production is required. Prior experience managing or leading engineering teams is a must. Experience building or operating agentic AI systems or complex multi-step AI workflows is highly preferred.

Technical proficiency should include Infrastructure-as-Code (Terraform, Pulumi, CloudFormation), Kubernetes, containerization (Docker), and CI/CD tooling. Familiarity with MLOps/LLMOps tooling, model-hosting economics, GPU resource management, and inference optimization techniques is expected. A working knowledge of enterprise security, data residency, and compliance requirements in regulated industries is also necessary. Proficiency in Python is strongly preferred, with Go or Java being a plus.

We value proven ability to lead engineering teams in fast-paced, ambiguous environments, strong cross-functional collaboration skills, and clear communication to translate technical trade-offs into business impact for leadership. Comfort operating in a "zero to one" environment, building foundational platforms, is key.