Principal ML Engineer — AI Agent Security & Identity

Okta; United States; Híbrido; USD 135 000 - 186 000; Tempo integral

Vantagens oferecidas por esta oferta de emprego

Equity (where applicable)

Bonus

Health insurance

401(k)

Resumo da oferta

Okta is seeking a Principal ML Engineer on the Secures AI team to advance our authorization capabilities. You will replace static rules with dynamic AI security, delivering real-time threat inspection and agent behavior analysis to keep autonomous agents safe at scale.

You will architect features for our unified control plane and mentor engineers, collaborating with product teams to push greenfield initiatives forward in a fast-paced, security-first environment.

Qualificações

• 10+ years of software development experience.
• Strong Python programming skills and ML experience.
• Hands-on ML with modern AI platforms (AWS Bedrock, OpenAI, Anthropic).

Responsabilidades

• Implement intent-based enforcement to verify agent runtime requests.
• Apply LLM reasoning to interpret prompts, tool payloads, and intent in real time.
• Integrate low-latency inference engines into the API gateway path.
• Use embeddings and vector search to assess alignment of agent actions.
• Design guardrails and evaluation benchmarks to prevent bypasses.
• Architect scalable ML systems integrating retrieval, inference, and evaluation pipelines.
• Mentor engineers and collaborate with product teams on greenfield AI initiatives.

Conhecimentos

Python

ML engineering

LLM platforms

RAG & embeddings

APIs & FastAPI

Team leadership

Security mindset

LangChain/LangGraph

Knowledge of AI safety

Formação académica

Bachelor’s or Master’s in Computer Science

Ferramentas

PyTorch

TensorFlow

LangChain

MCP

Airflow

Descrição da oferta de emprego

Okta is seeking a Principal ML Engineer on the Secures AI team to advance our authorization capabilities. You will replace static rules with dynamic AI security, delivering real-time threat inspection and agent behavior analysis to keep autonomous agents safe at scale.

You will architect features for our unified control plane and mentor engineers, collaborating with product teams to push greenfield initiatives forward in a fast-paced, security-first environment.