Staff Engineer, Agentic Intelligence - San Francisco

The role

You will own the policy framework and the agent harness for an AI underwriting platform. You will design systems to encode underwriting judgment into policies and build datasets to measure policy quality.

What you will do

• Own the policy framework.
• Design and build the framework that turns underwriting judgment into policies.
• Author policies that translate underwriting rules into testable policy.
• Establish and tune evals to measure policy behavior.
• Build datasets and harnesses to measure policy quality.
• Maintain the agent runtime including tool use, orchestration, context management, retries, and guardrails.
• Establish patterns, libraries, and review standards.
• Mentor and recruit engineers.
• Ship capabilities from ideas to production.
• Iterate rapidly based on real file performance.

Required qualifications

• Strong engineering proficiency.
• Ability to wield AI-powered coding tools.
• Experience building and operating LLM-based or agentic systems in production.
• Experience designing configuration-driven systems.
• A rigorous, empirical approach to evaluation.
• Comfort working across the stack with Go, Python, and TypeScript.
• Track record of leading a technical surface end-to-end.

Nice to have

• Worked on decisioning, document understanding, or human-in-the-loop products.

Benefits

• Competitive compensation calibrated to senior/principal-level engineers in San Francisco
• Meaningful equity ownership
• Full medical, dental, and vision coverage
• 401(k) with company match
• Flexible vacation policy
• Parental leave
• Learning and professional development support

About the company

HomeVision is the AI underwriting platform for the U.S. mortgage industry. Their platform, MIRA, automates appraisal reviews and is expanding to cover the full underwriting workflow.