Staff Engineer, Harness Engineering

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. We're adding ~1,000 customers per month. We've grown 100x since last year. We're scaling toward Series B.

Harper runs coding agents and product agents across two surfaces: product agents (intake, sales, service, voice - the AI that operates the business) and coding agents (frontier coding agents, OpenClaw, our internal Hermes-based agents - the AI that operates the codebase). Both ride on a shared substrate: the harness.

Own the meta-harness. Every Harper agent - product or coding - runs on what you build.

The Thesis

The companies that win with agents do not win because they found one magic model. They win because they build the harness that turns models into reliable workers: the loop, the tools, the memory, the execution environment, the budget, the fallback path, the guardrails. The model is replaceable. The harness is the company-specific leverage.

Harper needs that leverage everywhere: product agents, coding agents, voice agents, underwriting agents, service agents. You will own the layer they all run through. If you do this right, every new agent ships faster, fails smaller, and learns from the systems that came before it.

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.

Underneath all of that lives the harness substrate you'll own - the meta-harness wrapping our frontier coding agents, OpenClaw, Hermes, and the model-routing layer on top of our foundation-model providers. You define the contracts every Harper agent integrates against.

What You'll Own

• The agent loop - Prompt construction, tool routing, context-window management, retry/timeout/budget logic, multi-step orchestration

• Execution environment - Sandbox lifecycle, isolation, blast-radius limits, file-system + network policy for agents that browse or call APIs

• Tool layer - The canonical set of tools every agent can call. Schema, auth, rate-limit, observability per tool.

• Model-provider abstraction - Provider routing, fallback chains, cost/latency tradeoffs, eval-driven model selection

• Multi-agent coordination patterns - Parent/child handoff, shared memory contracts, parallel execution, conflict resolution, subagent budgets

• The harness SDK - What every pod engineer imports when they ship an agent. If pod engineers ship faster, you did this right.

• Guardrails - Banned tool combinations, prompt-injection defense, data-egress policy, PII scrubbing

You Might Be a Fit If…

• You've built or owned an agent harness at an AI-substrate company (or a strong open-source equivalent)

• You can describe trade-offs between agent loops you've used and have opinions on which works when

• You think in tool design, not just prompt design - the prompt is the last 5%, the tool surface is the work

• You've shipped sandbox infrastructure at scale (Firecracker, gVisor, comparable isolation)

• You write code with AI daily and have strong opinions about which harness behaviors matter and which are theater

• You're 8–12 years into your career with 3+ years at the Senior+ level

Requirements

• 8+ years software engineering experience, including senior+ scope at a high-growth company

• Production agent-harness or AI-substrate experience - agent loop, tool routing, execution environment, model routing

• Strong written communication - RFCs, API contracts, integration guides

• Based in San Francisco or willing to relocate

Nice to Have

• Open-source contribution to agent/harness frameworks

• Sandbox/isolation infrastructure depth

• Foundation-model partner or early-access experience

Compensation

• OTE: $253,000–$308,000 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

• Uber commuter benefits

• Meals provided — breakfast, lunch, and dinner on us

• Snacks, drinks, and coffee stocked daily

• Free gym membership

• Health, dental, and vision insurance

The Process

1. Founder call (15 min) - Mission, pace, scope

2. CTO deep-dive (60 min) - Harness architecture conversation

3. Super Day on-site - full-day simulation of working at Harper: harness design, code review, cross-functional sessions, and founder/CTO time

4. Founder + CTO offer conversation - No committee. Best offer, first.

To Apply

If you've built a harness other engineers couldn't live without - send your resume, the harness, and tell us about a model-routing or sandbox decision you'd defend in five years.