Member of Technical Staff, Evaluation Platform

Baseline Robotics; Zürich; Hybrid; CHF 110,000 - 170,000; Full time

Job summary

Baseline Robotics in Zürich seeks an engineer to design and operate the evaluation platform for our shared robotics hardware. You will implement scheduling, supervision, and scoring across evaluation runs to ensure fair comparisons and reproducible results.

You will also build interfaces for labs to submit policies and for reviewers to inspect results, while owning the data storage, access controls, and audit trails to maintain integrity and traceability.

Qualifications

• Experience running production systems where a wrong result is costly.
• Experience with evaluation infrastructure, benchmarks, or experiment tracking for machine learning.
• Understanding of how robot policies consume observations and emit actions under latency limits.

Responsibilities

• Design and build the scheduling, supervision, and scoring of evaluation runs on our stations.
• Design the interfaces for labs and agents to submit policies and review results.
• Own evaluation data storage, access control, and audit logging.

Skills

Production systems

Evaluation infrastructure

Machine learning experiments

Tools

Experiment tracking tools

Job description

Open source revolutionised how we write software, but its impact on robotics stays limited due to expensive hardware accessible only to well-funded labs. We run that hardware as shared infrastructure for the broader community, billed only for active station time. For a given task, policies are evaluated under similar conditions, giving researchers common ground for fair comparisons.

In this role you will:

• Design and build the scheduling, supervision, and scoring of evaluation runs on our stations.
• Design the interfaces for labs and agents to submit policies and review results.
• Own evaluation data storage, access control, and audit logging.

What we hope you’ll bring:

• Experience running production systems where a wrong result is costly.
• Experience with evaluation infrastructure, benchmarks, or experiment tracking for machine learning.
• Understanding of how robot policies consume observations and emit actions under latency limits.