Senior Software Engineer — Applied AI for Secure Identities

Okta; San Francisco (CA); On-site; USD 165,000 - 226,000; Full time

Benefits offered by this job

Health insurance

Dental insurance

Vision insurance

401(k)

Flexible spending account

Paid time off

Parental leave

Job summary

Okta in San Francisco seeks a Senior Software Engineer to help build Digital Workers—autonomous identity agents for secure operations. You will create harnesses for planning, tool use, memory, and safe execution across enterprise systems, integrating OpenAI and Anthropic APIs with a startup-like pace and the resources of a public company.

You will shape product quality and reliability, using evidence over conjecture, with a strong emphasis on security, evaluation, and scalable, maintainable

Qualifications

• Shipped AI software with tool-based workflows.
• Experience building agent-based evaluations for LLM apps.
• Strong ability to reason about model outputs and safety.

Responsibilities

• Build the agent harness for planning, tool use, memory, context compaction, interruption and recovery during long-running work.
• Integrate OpenAI and Anthropic model APIs with quality, latency, and cost considerations.
• Evaluate harnesses and implement missing pieces with AWS workflow integration.
• Enforce policy checks and collect evidence before task completion.
• Create evaluation datasets and replay tests for AI agents and access control.
• Quantify false positives, missed risks, unsafe actions and cost; derive regressions.
• Test for prompt injection, stale evidence, conflicting instructions and secure tool usage.

Skills

Python engineering

LLM app shipping

Agent workflows

Model API integration

Failure analysis

Tools

Claude Agent SDK

OpenAI Codex SDK

Strands Agents

LangChain Deep Agents

LangGraph

Job description

Okta in San Francisco seeks a Senior Software Engineer to help build Digital Workers—autonomous identity agents for secure operations. You will create harnesses for planning, tool use, memory, and safe execution across enterprise systems, integrating OpenAI and Anthropic APIs with a startup-like pace and the resources of a public company.

You will shape product quality and reliability, using evidence over conjecture, with a strong emphasis on security, evaluation, and scalable, maintainable