The Problem

36 million businesses in America need insurance - it's not optional. 77% are underinsured. 40% have no coverage at all. We're building 90%+ AI-led commercial insurance distribution. ~1,000 new customers/month, 100x growth in a year.

Every Harper agent - product and coding - runs on a shared substrate: the harness. Agent loop, tools, execution environment, model routing, multi-agent coordination, guardrails. If the harness is brittle, every pod ships brittle agents.

Build the substrate every Harper agent runs on. Make the harness so good that pod engineers ship agents in days, not weeks.

The Thesis

The model is only one part of the worker. The loop, tools, memory, execution environment, guardrails, fallback path, and observability determine whether an agent can actually do useful work inside a real company. That is the harness.

Harper needs harness engineers who want to build the substrate beneath every product agent and coding agent we ship. If you do this right, pod engineers stop rebuilding the same scaffolding and start shipping agents in days.

The Role

Harper operates like a factory with a series of modules spanning the full lifecycle from intake through renewals. Across them we run a stack of internal AI systems covering operator guidance, the operational backbone that matches risks to underwriters, autonomous communications, and voice AI for customer interactions.

You build inside the harness layer - the meta-harness wrapping our frontier coding agents, OpenClaw, Hermes, and our internal agents. You'll work alongside the engineer setting the harness direction and ship the primitives every agent team depends on.

What You'll Do

Build agent loop primitives - Prompt construction, tool routing, retry/timeout/budget logic, multi-step orchestration
Ship execution-environment infra - Sandbox lifecycle, isolation, blast-radius limits, filesystem + network policy for agents
Own the tool layer for assigned domains - Schema, auth, rate-limit, observability per tool. Tools other engineers consume.
Wire model-provider routing - Provider fallback chains, eval-driven model selection, cost/latency tradeoffs
Ship harness SDK improvements - Whatever makes pod engineers faster
Eat your own dog food - You write code with our harness daily and feel every rough edge

You Might Be a Fit If…

You've shipped production agentic systems (not demos - real users, real traffic)
You've worked with at least one major agent framework
You can describe a tool-design or sandbox decision you'd defend in three years
You write code with AI daily and manage 3+ parallel sessions
You're 3–6 years into your career

Requirements

3–6 years software engineering experience
Production agent / LLM systems experience - agent loops, tool integration, prompt engineering at scale
Strong written communication - API contracts, integration guides, internal docs
Based in San Francisco or willing to relocate

Nice to Have

Sandbox/isolation infrastructure experience
Open-source contributions to agent frameworks
Foundation-model partner or early-access experience

Compensation

OTE: $187,000–$264,000 cash compensation (base salary + target performance bonus)
Equity: competitive equity, so you share in the company you are helping build
Location: San Francisco, in-office

Benefits

Health, dental, and vision insurance
Commuter benefits
Team meals and snacks

The Process

Founder call (15 min) - Mission, pace, scope
Tech Lead deep-dive (60 min) - Harness conversation
Super Day on-site - full-day simulation of working at Harper: harness design, code review, team context, and founder/CTO time

Senior Member of Technical Staff, Harness Engineering