Member of the Technical Staff - Agent Infrastructure

Probook•New York, NY

4d•$200,000 - $300,000

About The Position

Probook is building an AI operating system for home services, with customers generating $10M to $100M in annual revenue and running their operations on the platform. The Member of the Technical Staff - Agent Infrastructure will build and operate the infrastructure underpinning Probook’s real-time decision systems, agent workflows, and developer platform. The role owns product infrastructure such as durable execution, workflow orchestration, reliability, and performance; agent sandboxes, evaluation infrastructure, and harnesses, including compute environments; and developer tools for CI/CD, deployment, and local development. It also encompasses observability, incident response, debugging, cloud and inference cost management, capacity planning, and architecture decisions. The engineer will partner closely with product engineers and customers affected by the infrastructure.

Requirements

3–6 years of engineering experience, with meaningful ownership of infrastructure built rather than inherited.
Zero-to-one experience building infrastructure from scratch at a startup or standing up new systems within a larger company.
Strong infrastructure fundamentals across distributed systems, cloud infrastructure, networking, CI/CD, and observability.
Strong software engineering skills, including writing code and owning complex systems end to end, from design through production.
Breadth across the stack and a preference for building new systems over operating mature ones.
Sound judgment about when a system needs careful architecture and when it needs to be finished quickly.

Responsibilities

Build and operate infrastructure for real-time decision systems, including durable execution, workflow orchestration, reliability, and performance.
Build sandboxes, evaluation infrastructure, and harnesses for developing and shipping agent workflows, including the compute environments in which agents execute.
Build and maintain the developer platform, including CI/CD, deployment, local development, and internal tooling.
Develop instrumentation, monitoring, incident response, and debugging tools for systems with probabilistic failure modes.
Manage cloud and inference costs as the product grows, and make capacity and architecture decisions before bottlenecks arise.
Partner with product engineers building on the infrastructure and with customers affected by it.
Own complex infrastructure systems end to end, from design through production.