AI Operations Engineer

Meet the Team Join Cisco's Commerce Intelligence Data & Analytics team, a pivotal group delivering seamless intelligence, advanced analytics, and cutting-edge agentic experiences across our business operations. Our mission is to build a scalable, unified, and AI-ready data foundation that drives high-impact business decisions and automated Agentic actions . As a specialized team of engineers and operational experts, we move beyond traditional maintenance to solve complex, large-scale commerce challenges with agentic AI. We achieve this by blending innovation, deep process knowledge, technical expertise, and a relentless focus on business impact, ultimately enhancing operational efficiency and data-driven decision-making at scale.

As an AI Engineer, you will design, build, and operate the autonomous and semi-autonomous agents that power Cisco's commerce operations. You will translate ambiguous operational requirements into reliable multi-step agent workflows — orchestrating LLM reasoning, tool and function calling, retrieval, and integrations into commerce platforms and operational systems. You will be responsible for the full lifecycle: designing the agent and its guardrails, grounding it with retrieval, connecting it to enterprise systems through robust API and agent-to-agent (A2A) integrations, evaluating its behavior, and deploying and monitoring it in production by collaborating with extended team across various time zones. You will partner with cross-functional AI teams to identify high-value opportunities and deliver intelligent, automated workflows that optimize performance and drive durable growth. This role is a pillar to our operational strategy and to Cisco's AI-first vision.

Architect and build production AI agents — including multi-agent and human-in-the-loop workflows — that automate commerce operations. Work in partnership and guidance from the core team at onsite Implement agent orchestration with modern frameworks (e.g., LangGraph, LangChain, AutoGen, CrewAI, or equivalent), designing robust tool/function-calling, planning, and state management. Build agent-to-agent (A2A) integrations and interoperability using emerging standards and protocols (e.g., A2A, MCP), enabling agents to delegate, coordinate, and hand off work reliably. Design and implement API integrations — REST/GraphQL, event-driven and webhook-based services, authentication, rate limiting, and resilient error handling — to connect agents with commerce platforms and enterprise data sources. Manage the end-to-end lifecycle of LLM-powered applications using Snowflake Cortex, from model selection and fine-tuning to deployment, versioning, and decommissioning. Architect robust agentic workflows using Snowflake Native Apps and Stored Procedures, implementing sophisticated tool-calling, planning, and state management that ensures seamless execution across complex operational tasks. Build and tune retrieval-augmented generation (RAG) pipelines over documents, structured data, and knowledge repositories. Implement and maintain low-latency vector databases, semantic search architectures, and Retrieval-Augmented Generation (RAG) pipelines using native vector data types and search functions. Build automated monitoring mechanisms to track LLM drift, hallucination rates, and token usage, triggering self-healing alerts and automated model retraining or routing workflows. Proactively manage and optimize Snowflake credit consumption for AI workloads; implement warehouse auto-scaling, query profiling, and multi-cluster strategies to ensure cost-efficiency at scale. Design guardrails and human-in-the-loop gates for sensitive actions, and implement LLM routing and failover for reliability and cost control. Apply prompt engineering and context engineering to make agent behavior accurate, safe, and consistent under real operational conditions. Own the AI/ML CI/CD pipeline, ensuring rigorous version control,