Principal ML Engineer: AI Security & Policy Orchestration

Okta

San Francisco (CA)

Hybrid

USD 238,000 - 326,000

Full time

6 days ago

Be an early applicant

Application generator

Stand out for this role — generate a tailored resume and cover letter in about a minute.

Get past ATS filters

Benefits offered by this job

Equity

401(k)

Flexible spending account

Paid leave

Job summary

Okta is seeking a Principal Machine Learning Engineer to evolve authorization for agent access policies using dynamic AI security mechanisms. You will design scalable ML and Generative AI systems with retrieval, inference, and evaluation pipelines, while mentoring engineers and collaborating with product and security teams to enhance safety and reliability of AI-powered access controls.

The role requires 10+ years of software development with Python, deep ML experience, and hands-on work with

Qualifications

10+ years of software development experience with Python.
Hands-on experience with applied machine learning, including feature engineering, training, and fine-tuning models.
Experience with modern Generative AI platforms such as AWS Bedrock, OpenAI, Anthropic.
Deep understanding of retrieval-augmented generation (RAG), embeddings, and knowledge-base workflows.
Hands-on experience with AI agent frameworks including LiteLLM, LangGraph, LangChain, LlamaIndex, MCP.

Responsibilities

Build intent-based enforcement to confirm agent runtime requests match intended purpose.
Apply LLM reasoning and prompt parsing to interpret prompts, tool payloads, and intent in real time.
Integrate low-latency inference or semantic evaluation engines into the API gateway request path.
Use embeddings, vector search, or zero-shot classification to score alignment between agent intent and executed actions.
Design confidence-scored decision engines that feed semantic verification results into Cedar policy frameworks.
Establish evaluation benchmarks, prompt injection defenses, and guardrails to prevent bypasses or false positives.
Architect scalable ML and Generative AI systems with retrieval, inference, and evaluation pipelines.
Optimize prompting, context retrieval, and RAG workflows for Claude-based systems.
Build automated evaluation pipelines to measure model quality, correctness, groundedness, and safety in production.
Implement schema validation, structured output enforcement, and guardrails for reliable, compliant AI outputs.
Mentor and coach engineers to support team and community growth.