Autonomy Evaluation Tech Lead — Metrics & Scale

Atoms; San Francisco (CA); On-site; USD 185,000 - 242,000; Full time

Benefits offered by this job

Medical, Dental, Vision insurance

Equity

Paid time off

Job summary

Atoms is building machines that power the next era of progress in San Francisco. We seek a Technical Lead for Autonomy Evaluation to own the metrics our autonomy releases are judged on and the log replay and evaluation platform that computes them.

You will take evaluation from recorded logs to scored regression suites for every release. You will develop how replay, scoring, metrics, and test coverage come together into a single platform used daily by engineers, across on-road vehicles.

Qualifications

• Bachelor's degree in CS/Robotics/EE/Statistics with 8+ years of relevant experience.
• Experience leading evaluation or test teams for autonomous systems or large-scale ML evaluation.
• Hands-on experience building log replay at scale with scoring against ground truth.

Responsibilities

• Own safety and behavior metrics and go/no-go criteria for releases.
• Build and own the log replay and scoring pipeline across perception, localization, and planning.
• Benchmark software across logs and design evaluation datasets for regression detection.
• Ensure deterministic, reproducible evaluation with traceability to software versions.
• Mentor engineers, define testing practices, and participate in hiring.
• Collaborate with cross-functional teams to advance evaluation infrastructure.

Skills

Python

C++

Statistics

Log replay

GPU compute

Education

Bachelor's degree in Computer Science, Robotics, Electrical Engineering, Statistics, or a related field

Tools

GPU clusters

Log replay tooling

Robotics middleware

Job description

Atoms is building machines that power the next era of progress in San Francisco. We seek a Technical Lead for Autonomy Evaluation to own the metrics our autonomy releases are judged on and the log replay and evaluation platform that computes them.

You will take evaluation from recorded logs to scored regression suites for every release. You will develop how replay, scoring, metrics, and test coverage come together into a single platform used daily by engineers, across on-road vehicles.