About the Role
The AI Operations Engineer is responsible for building the central knowledge base for the company. You will bridge the gap between our raw data and our AI assistants. You will ensure our teams can use the world’s best models (Gemini, Claude, ChatGPT) safely and efficiently. Your mission is to design secure RAG (Retrieval-Augmented Generation) structures that allow AI to safely access company knowledge while strictly respecting security boundaries and data privacy.
This role is onsite 5 days a week at our Mountain View, CA office!
What you'll do
AI Stack Governance: Administer our AI suite, including Gemini CLI and Claude Code for engineers, and Granola AI for general productivity.
Secure RAG Architecture: Design and maintain the vector databases and data pipelines that power internal AI. Ensure the AI only "sees" what the user’s Google/Okta permissions allow.
Agentic Workflows: Build and tune AI agents using frameworks like LangChain, LlamaIndex and Zapier to automate complex business processes.
Governance-Before-Generation: Implement "Identity-Aware Retrieval" to ensure AI outputs never leak sensitive data to unauthorized users.
AI Security (AI-SPM): Protect the organization against AI-specific threats like prompt injection and data poisoning. Monitor for "Shadow AI" usage.
Performance Tuning: Monitor LLM latency, token usage, and accuracy, optimizing embeddings and "chunking" strategies for better AI performance.
What we're looking for
Proven experience administering Enterprise AI platforms (Gemini, Claude, ChatGPT).
Strong Python skills for building RAG pipelines and interacting with LLM APIs.
Familiarity with Vector Databases (e.g., Pinecone) and AI orchestration (LangChain/LlamaIndex).
Understanding of the security intersection between IAM (Google/Okta) and AI dataretrieval.
Familiarity with data privacy standards and how to mask PII within AI prompts