Machine Learning Software Engineer

About SafeWorld

SafeWorld builds robot safety testing and simulation software for teams deploying autonomous systems around people.

Our mission is to improve robot safety for the next billion machines.

Build with Ding Zhao, Kyle Wong, and Simo Rachidi.

Ding has spent nearly two decades working on the safety of autonomous systems.

He leads Carnegie Mellon’s Safe AI Lab and is a National Science Foundation CAREER Award recipient.

We’re already working with robotics companies and global enterprises.

We recently announced $12. 2M in seed funding, co-led by Shine Capital and a16z speedrun.

The role

Build the machine learning systems behind repeatable robot safety tests.

Turn robot logs, incidents, and simulation results into useful models, measurements, and engineering decisions.

Work directly with the founders and customers to turn research into reliable software.

What you’ll do

• Build models for risk scoring, near-miss detection, and scenario ranking.
• Own data ingestion, labeling, training, evaluation, and deployment.
• Build reliable pipelines for robot logs, time series, events, and simulation results.
• Turn safety requirements into clear model targets, metrics, and evaluation reports.
• Test models against real incidents and varied simulated conditions.
• Add monitoring and drift detection to maintain model quality.
• Build production Python systems with frameworks such as PyTorch, JAX, or TensorFlow.
• Help set engineering standards through code reviews, tests, documentation, and customer feedback.

What you bring

• Experience shipping and maintaining production Python systems.
• Experience training and evaluating machine learning models on real data.
• Strong skills in data pipelines, debugging, and software design.
• Clear communication and comfort taking ownership in a small team.
• Curiosity about robotics and care for clear evidence in safety decisions.

Explore our work and current roles: https://www. safeworld. ai/careers