Senior Full-Stack Engineer, Autonomous Defense

horizon3ai; United States; Hybrid; USD 120,000 - 170,000; Full time

Benefits offered by this job

Stock options

Health, vision, and dental coverage

Flexible vacation

Parental leave

Job summary

horizon3ai is building the platform that runs autonomous defensive agents across customer environments. You will own backend systems and operator-facing interfaces, ensuring durable, observable, and safe agent workflows at scale.

Focus areas include execution, orchestration, integrations, data processing, and a production-first mindset. You will work with AI researchers while maintaining reliable systems across many environments.

Qualifications

• Must have strong backend engineering experience with distributed systems.
• Ability to deliver frontend features as part of end-to-end products.
• Capable of building reliable systems across many customer environments.
• Understand how to make actions controlled, observable, and auditable.
• Experience working with AI researchers while supporting production systems.

Responsibilities

• Develop a runtime for long-running agent workflows with checkpoints, resumable steps, timeouts, and cancellation.
• Create a versioned tool registry with schema checks and server-side authorization.
• Build connectors to security systems (endpoint protection, firewalls, identity, SIEM, cloud access).
• Implement safeguards: simulation, impact assessment, approvals, staged rollout, rollback, auditable records.
• Build operator interfaces in React and TypeScript with live WebSocket updates.
• Strengthen tenant isolation with execution boundaries, egress controls, quotas, and data leakage protection.

Skills

Backend engineering

Distributed systems

Frontend features

Reliability across environments

Observability & auditable

Collaboration with AI researchers

Tools

React

TypeScript

WebSockets

Job description

Role overview

Build the platform that runs autonomous defensive agents across customer environments. You’ll own backend systems and operator-facing interfaces, making agent workflows durable, controlled, observable, and safe to use at scale. The work spans execution, orchestration, integrations, data processing, and user interfaces; the focus is production infrastructure rather than tuning AI prompts.

Responsibilities

• Develop a runtime for long-running agent workflows, including checkpoints, resumable steps, cancellation, idempotency, and timeouts.
• Create a versioned tool registry with schema checks, tenant-level controls, and server-side authorization.
• Build connectors to security systems such as endpoint protection, firewalls, identity, SIEM, and cloud access management.
• Implement safeguards for proposed changes, including simulation, impact assessment, approvals, staged rollout, rollback, and auditable records.
• Build operator interfaces in React and TypeScript, with live updates over WebSockets so users can inspect and manage ongoing runs.
• Strengthen tenant isolation for agent actions, including execution boundaries, egress controls, quotas, and protection against cross-tenant data leakage.

Requirements

• Strong backend engineering experience, especially with distributed systems and production platforms.
• Comfort delivering front-end features as part of end-to-end product work.
• Ability to build systems that remain reliable across many customer environments and long-running workflows.
• Understanding of how to make automated actions controlled, observable, and auditable.
• Ability to work closely with AI researchers while owning the production systems that support their work.

Benefits and work setup

The role is described as remote or hybrid depending on location. Full-time roles include stock options; listed benefits include health, vision, and dental coverage, flexible vacation, and parental leave.