AI Platform Engineer

Role overview

Help take language-model and agent-based applications from early prototypes into reliable production services. The work spans application development, cloud infrastructure, deployment, monitoring, and cost management, with an opportunity to improve the technical practices used to build AI systems across the organization.

Responsibilities

- Design, build, and deploy production LLM applications, including retrieval-augmented generation pipelines, agents, and workflow automation. - Assess models, frameworks, and providers for each use case, considering hosted and open-source options. - Create evaluations and benchmarks to check output quality, reliability, and regressions before and after releases. - Build and manage cloud infrastructure for AI workloads using a major cloud platform and infrastructure-as-code practices. - Maintain CI/CD workflows for models, prompts, and AI services, including automated checks, controlled rollout, and rollback. - Deploy and scale AI services with containers and orchestration tools; Docker and Kubernetes are named examples.

Requirements

- The role calls for the ability to take AI solutions through architecture, development, deployment, monitoring, and cost optimization. - Practical knowledge of LLM application patterns such as RAG, agents, or automated workflows is relevant to the work. - Familiarity with cloud infrastructure and infrastructure as code is needed for operating AI workloads. - Experience with containerized deployment and orchestration is relevant; Docker and Kubernetes are cited examples. - Ability to build evaluation and release processes that help assess quality and reliability is required by the described responsibilities.