Safety Systems Engineer — Full-Stack Tools for Safe AI

United States Digital Space LLC; United States; Remote; USD 120,000 - 190,000; Full time

Job summary

United States Digital Space LLC is seeking an experienced software engineer to build internal tooling that enhances the safety and reliability of our AI models. You will own full-stack development, collaborating with researchers, engineers, and policy writers to turn safety ideas into durable product solutions.

The role focuses on areas like labeling data, inspecting failure cases, and surfacing insights for iteration, with an emphasis on speed, usability, and scale to accelerate safety

Qualifications

• 5+ years of engineering experience at tech/product-driven companies.
• Proficiency with JavaScript and React.
• Proficiency with at least one backend language (Python).
• Experience with relational databases like Postgres/MySQL.
• Interest in AI/ML and collaboration with researchers and policy writers.

Responsibilities

• Own end-to-end development of internal tools to improve model safety.
• Partner with Safety Systems researchers, engineers, and policy creators to translate workflows into durable tools.
• Build full-stack experiences for core model policy workflows (labelling, inspecting data, analyzing failures).
• Optimize internal apps for usability, speed, and scale to boost velocity and time-to-fix for safety issues.
• Transform AI-assisted safety workflows into external-facing safety products for safer AI.

Skills

JavaScript

React

Python

Relational databases

AI/ML interest

Job description

United States Digital Space LLC is seeking an experienced software engineer to build internal tooling that enhances the safety and reliability of our AI models. You will own full-stack development, collaborating with researchers, engineers, and policy writers to turn safety ideas into durable product solutions.

The role focuses on areas like labeling data, inspecting failure cases, and surfacing insights for iteration, with an emphasis on speed, usability, and scale to accelerate safety