Requirements Bachelor’s or Master’s degree in Computer Science, Software Engineering, or a related field 2+ years of professional experience in Platform Engineering, DevOps, SRE, or backend software development Strong experience and confidence working with cloud infrastructure (AWS preferred) and Infrastructure as Code (Terraform) Proficiency in at least one programming language (e.g., Python, Go) and experience with CI/CD pipelines and containerization (e.g., Docker) Solid understanding of API cost management, rate limiting, security best practices, and data privacy for AI models Experience with, or a strong willingness to learn, LLM gateways, observability platforms, and telemetry (e.g., LiteLLM, Langfuse, Datadog) Strong problem-solving skills, a sense of humor, and the ability to thrive in an ego-free environment as an embedded platform expert within a cross-functional Data Science and AI team What the job involves As an AI Platform Engineer embedded within our Data Science and AI tribe, you will design and maintain the foundational infrastructure that enables our AI Software Engineers to build and scale generative AI applications You will build our internal AI provisioning platform, focusing on seamless model access, observability, and strict governance In this role, you will ensure our AI integrations are secure, performant, and cost-effective, providing the robust backbone required for our agentic and LLM-driven workflows Develop and maintain an internal AI provisioning platform to streamline access to various LLMs & MCP Servers (e.g., via LiteLLM) Implement comprehensive LLM observability, logging, and alerting systems using tools like Datadog and Langfuse Enforce model governance, ensuring security, data privacy, rate limiting, and strict LLM API cost management Collaborate closely with embedded AI Software Engineers to support cloud-native architectures and agentic workflow integrations Build and maintain infrastructure for monitoring the reliability and performance of production AI systems #J-18808-Ljbffr
