Staff AI Engineer

HartleyCo; San Francisco (CA); On-site; USD 300,000 - 400,000; Full time

Job summary

HartleyCo in San Francisco is seeking a Staff AI Engineer for the Agent Team to design and build production-grade AI agents that automate complex enterprise workflows. You will operate above foundation models, drive architecture decisions, and shape the product direction for a category-leading company.

The role requires 8+ years in software engineering with 3+ years in AI or ML, strong Python/backend skills, and hands-on experience shipping production LLM or agent systems.

Qualifications

• 8+ years of software engineering experience.
• 3+ years building in AI or machine learning.
• Strong Python and backend fundamentals.
• Recent experience shipping production LLM or agent systems.
• Ownership of architecture beyond individual features.

Responsibilities

• End-to-end evaluation systems for agent performance.
• Agent orchestration across long-running workflows.
• Tools and MCP servers connecting agents to enterprise systems.
• Context, memory and reasoning strategies.
• Multi-agent systems and feedback-driven improvement loops.
• Production reliability, observability and failure recovery.
• Experiments that turn emerging AI research into product capabilities.

Skills

Python

Backend engineering

AI/ML experience

LLM/agent systems

Production reliability

Job description

I’m supporting one of the fastest-growing applied AI companies in San Francisco as they hire a Staff AI Engineer for their Agent Team.

The company is building production AI agents for the insurance industry, replacing complex manual workflows across carriers, brokers and enterprise businesses.

They have:

• Raised more than $30M from a16z, Y Combinator, Nexus and South Park Commons
• Grown rapidly over the past 12 months
• Signed top-10 insurance carriers and brokers
• Deployed with Fortune 100 and public companies
• Built a deeply technical team with experience from Apple AI Research, Brex, Coinbase, Uber and multiple YC startups

This is not a model-training or research role.

You’ll be building the systems above the foundation models that make agents reliable in production.

What You’ll Own

• End-to-end evaluation systems for agent performance
• Agent orchestration across long-running workflows
• Tools and MCP servers connecting agents to enterprise systems
• Context, memory and reasoning strategies
• Multi-agent systems and feedback-driven improvement loops
• Production reliability, observability and failure recovery
• Experiments that turn emerging AI research into product capabilities

Both founders are engineers, and engineers have a genuine say in product direction.

This is a hands-on Staff role for someone who wants to set technical direction, make meaningful architecture decisions and help define how a category-leading company builds agents.

The Ideal Background

• 8+ years of software engineering experience
• 3+ years building in AI or machine learning
• Strong Python and backend engineering fundamentals
• Recent experience shipping production LLM or agent systems
• Ownership of architecture beyond individual features
• Experience with evals, orchestration, tool use, context engineering or agent reliability
• Strong communication and product judgement
• Comfortable working at speed in an ambitious, on-site environment

Previous startup, early engineering or founder experience would be particularly valuable.

Insurance experience is not required.

Compensation:

$300K-$400K base + meaningful equity