Okta Verified

Software, Cybersecurity, Identity Management

Principal Software Engineer, Applied AI

San Francisco, CAOnsiteFull TimePrincipal$224,000–$308,000 /yrPosted 22 days agoVisa sponsorship available

Is this role right for you?

Upload your resume and get a skill-by-skill breakdown — see exactly where you match, where you're close, and what to highlight. Not a mystery percentage.

Get a tailored resume highlighting what this role needs.

Required skills

AI software
agent execution
model quality
Python
LLM application
agent workflows
evaluation systems
tool-using LLM applications
Claude Agent SDK
OpenAI Codex SDK

Nice to have

Open-source contributions
agent observability
synthetic evaluation data
fine-tuning
preference data
post-training evaluation

Role summary

Okta is seeking a Principal Software Engineer to join their Applied AI team, focusing on building Digital Workers for identity operations and security. This role involves developing agent harnesses, integrating AI models (OpenAI, Anthropic), and establishing robust evaluation systems to ensure reliability and trustworthiness for Okta's customers. You will leverage strong Python skills, experience shipping LLM applications, and expertise in agent workflows or evaluation systems. The position offers the opportunity to shape product direction and implement shared components within a dynamic, well-resourced environment. Experience with AWS, PostgreSQL, and API integrations is crucial, with preferred experience in data engineering and security workflows.

Secure Every Identity, from AI to HumanIdentity is the key to unlocking the potential of AI. Okta secures AI by building the trusted, neutral infrastructure that enables organizations to safely embrace this new era. This work requires a relentless drive to solve complex challenges with real-world stakes. We are looking for builders and owners who operate with speed and urgency and execute with excellence.This is an opportunity to do career-defining work. We're all in on this mission. If you are too, let's talk.

The Principal Software Engineer Opportunity

The team that built Jeeva AI is building Digital Workers, a new suite of autonomous agents for identity operations and security inside Okta. The team works directly with Todd McKinnon, Okta's co-founder and CEO, with the opportunity to bring this product to Okta's 20,000+ customers. You will help shape the product with the ownership and pace of a startup and the resources of a public company.

You will build the harnesses and evaluations that make these workers reliable. They investigate risk, use tools, wait for approvals and complete tasks across enterprise systems. Bring depth in agent execution or model quality and help turn it into working software customers can trust.

About You

Production builder: You have shipped AI software and know what changes when real users depend on it.

Careful evaluator: You investigate failures and use evidence to judge quality.

Strong owner: You follow an improvement through code, evaluation, release and customer use.

What you will be doing

Build the agent harness for planning, tool use, memory, context compaction, interruption and recovery during long-running work.

Integrate OpenAI and Anthropic model APIs. Select models using measured quality, latency and cost, with clear failure handling.

Evaluate existing harnesses and build only the missing pieces. Keep the harness connected to AWS workflow execution and approval state.

Keep authorization and sensitive writes behind application policy checks. Require evidence of the resulting state before marking work complete.

Build evaluation datasets and replay tests for valid findings, access decisions, app setup and AI agent onboarding.

Measure false positives, missed risks, unsafe actions, completion rate and cost. Turn production failures into regression tests.

Test prompt injection, stale evidence, conflicting instructions and unsafe tool use. Protect tenant boundaries in retrieval and memory.

Improve prompts, retrieval, tools and model routing. Evaluate fine-tuning or other post-training methods when results justify them and data use is approved.

What you will bring to the role

Strong Python engineering skills and experience shipping an LLM application that uses tools.

Hands-on experience building agent workflows or evaluation systems for tool-using LLM applications. For harness work, relevant experience includes Claude Agent SDK, OpenAI Codex SDK, Strands Agents, LangChain Deep Agents, LangGraph or a custom implementation.

Experience integrating model APIs, structured outputs and asynchronous services. Experience with OpenAI or Anthropic is especially relevant.

Depth in agent execution or evaluation systems, with working code and failure analysis you can explain.

Ability to distinguish a plausible model response from a supported finding or verified action.

Scope by level

Principal: Set product engineering direction across workflows and implement the shared components that help the team ship faster.

Our stack

Our core stack is Python FastAPI, AWS Step Functions, Amazon Bedrock AgentCore, PostgreSQL on Amazon RDS and pgvector. OpenAI and Anthropic model APIs are part of the integration plan.

We plan to add an agent harness and a dedicated analytics layer using ClickHouse or BigQuery. The harness choice is open. Experience with equivalent tools also counts.

And extra credit for

Open-source contributions, agent observability, security evaluations or synthetic evaluation data.

Fine-tuning, preference data, post-training evaluation or model distillation using approved datasets.

Identity and security workflows. Foundation-model research experience is optional.

P26074

#LI-Inperson

#LI- BB1

Below is the annual base salary range for candidates located in San Francisco Bay Area. Your actual base salary will depend on factors such as your skills, qualifications, experience, and work location. In addition, Okta offers equity (where applicable), bonus, and benefits, including health, dental and vision insurance, 401(k), flexible spending account, and paid leave (including PTO and parental leave) in accordance with our applicable plans and policies. To learn more about our Total Rewards program please visit: https://rewards.okta.com/us.

The annual base salary range for this position for candidates located in the San Francisco Bay area is between: $224,000—$308,000 USD

The Okta Experience

Supporting Your Well-Being

Driving Social Impact

Developing Talent and Fostering Connection + Community

We are intentional about connection. Our global community, spanning over 20 offices worldwide, is united by a drive to innovate. Your journey begins with an immersive, in-person onboarding experience designed to accelerate your impact and connect you to our mission and team from day one.Okta is an Equal Opportunity Employer. All qualified applicants will receive consideration for employment without regard to race, color, religion, sex, sexual orientation, gender identity, national origin, ancestry, marital status, age, physical or mental disability, or status as a protected veteran. We also consider for employment qualified applicants with arrest and convictions records, consistent with applicable laws.If reasonable accommodation is needed to complete any part of the job application, interview process, or onboarding please use this Form to request an accommodation.Notice for New York City Applicants & Employees: Okta may use Automated Employment Decision Tools (AEDT), as defined by New York City Local Law 144, that use artificial intelligence, machine learning, or other automated processes to assist in our recruitment and hiring process. In accordance with NYC Local Law 144, if you are an applicant or employee residing in New York City, please click here to view our full NYC AEDT Notice.

Sample Okta interview questions

1
Develop a real-time platform for audio/video streaming.
system designmedium
2
Design a system for managing inventory in a warehouse.
system designmedium
3
Design a real-time system to track and monitor traffic flow.
system designmedium
4
Valid Word Abbreviation Determine if a string matches a valid word abbreviation. Input: word = "apple", abbr = "a2e" Output: FALSE Explanation: The abbreviation claims exactly 2 characters are skipped between 'a' and 'e', but "ppl" is actually 3 characters long.
codingmedium
5
Implement String Compression Implement string compression. Input: chars = ["a","b","b","b","b","b","b","b","b","b","b","b","b"] Output: ["a","b","1","2"] Explanation: The letter 'a' appears once (so no number is appended), and 'b' appears 12 times, modifying the array in place to length 4.
codingmedium

Get started to unlock a personalized interview prep pack for this role.

Ready to apply?

You'll be redirected to Okta's application page.