Job Purpose and Impact The Senior AI Platform Engineer for AI Assistants Agents in AI Data Science builds services to operate Cargill s portfolio of GenAI Assistants and Agents - including ChatGPT-powered copilots and bespoke LLM services You will translate product requirements into secure efficient and observable services and or agents manage the lifecycle from experiment to production and continually improve reliability latency and cost Typical deliverables span REST services vector-search back end LLMOps AgentOps pipelines and rich front-end components for knowledge workers and plant operators Expect to collaborate daily with product managers data scientists cloud engineers and security teams while acting as the technical authority for GenAI feature design and launch Key Accountabilities SOFTWARE DEVELOPMENT Designs and develops high quality software solutions by writing clean maintainable and effective codes AUTOMATION Leads the application of internal software deployment platform methodologies and tools to automate the deployment process ensuring smooth and reliable releases COLLABORATION Partners with cross functional team of product managers designers and different engineers to gather complex requirements and deliver solutions that meet business needs TESTING DEBUGGING Writes and maintains complex unit tests and integration tests and performs debugging to maintain the quality and performance of the software CONTINUOUS IMPROVEMENT Suggests options for improving the software development and deployment processes and implements the approved standards to improve efficiency and reliability DOCUMENTATION Builds and maintains comprehensive documentation for complex software applications deployment processes and system configurations TECHNICAL SUPPORT Provides technical support and troubleshooting for complex issues with deployed applications to ensure minimal downtime and fast resolution ESSENTIAL FUNCTIONS Design Build Develop multi-agent workflow automation patterns using Agentic AI Process redesign and mapping to agentic workflow patterns Architect scalable micro-services that wrap LLM RAG Agent workflows Python Implement robust prompt-engineering patterns retrieval pipelines and caching for AI Assistants and AI Agents Platform Ops Extend evaluation automated testing canary rollout and rollback for AgentOps Profile inference latency GPU CPU utilization and memory deliver quarterly cost-to-serve reductions Lead bug-fix security-patch and performance-tuning sprints for live AI Assistants and AI Agents Operational Excellence Own on-call runbooks SLOs and incident reviews embed observability Enforce Responsible-AI guardrails data-privacy controls and vulnerability-management policies for AI Assistants and AI agents Enablement Mentoring Coach full-stack and data-science peers on GenAI LLMOps patterns create internal workshops and tech blogs Qualifications Minimum 4 years building production software or data platforms Typical 5-8 years including 2 years with cloud-native AI ML or GenAI systems Azure AWS or GCP or 2 years of software devlopment