Principal Machine Learning Engineer, SecureAI

Okta is building the first Identity Security Fabric for the agentic era. As a Principal ML Engineer on the SecureAI team, you will architect and implement dynamic AI security mechanisms for real-time threat inspection, agent intent evaluation, and behavioral analysis, ensuring autonomous AI agents operate safely within specified bounds. You will apply LLM reasoning, embeddings, and RAG to build a unified control plane for identity management.

10+ years software development with Python; hands-on ML from feature engineering to fine-tuning; experience with GenAI platforms (AWS Bedrock, OpenAI, Anthropic); deep understanding of RAG, embeddings, and knowledge-base workflows; experience with AI agent frameworks (LiteLLM, LangGraph, LangChain, LlamaIndex, MCP); familiarity with ML frameworks (FastAPI, PyTorch, TensorFlow) and orchestration (Airflow); ability to define evaluation metrics and pipelines; proven collaboration on greenfield ini