Research Engineer / Data Scientist: Data-Centric AI & Model Evaluation

← Back to Careers

Research Engineer / Data Scientist: Data-Centric AI & Model Evaluation

Seoul Gangnam (On-site)

• Full-time·Talent Pool / Year-round Interest

About The Role

For a robot, a single average score can hide the failures that matter most. We are building a talent pool of data-centric AI and evaluation specialists who can turn robot behavior into trustworthy evidence: where a model works, where it fails, why it fails, and what data or algorithmic change is most likely to help.

You will define task taxonomies, benchmarks, metrics, annotation strategies, and experiment designs across perception, robot learning, VLA, world models, and end-to-end autonomy. You will connect offline model results, simulation outcomes, and real-robot runs into a decision system for research priorities and release quality.

Key Areas

Evaluation designFailure taxonomyData qualityExperimentationBenchmarkingError analysisRelease criteria

What You'll Do

• Define task, scenario, and failure taxonomies with metrics that reflect safety, reliability, efficiency, and recovery.
• Build evaluation datasets and harnesses spanning offline inference, simulation, replay, hardware-in-the-loop, and real-robot runs.
• Analyze slice-level performance, causal drivers, regressions, calibration, and statistical uncertainty.
• Design annotation guidelines, sampling strategies, active-learning loops, and data-quality audits.
• Create scorecards and release gates that help researchers and engineering leaders choose the next experiment and decide when a model is ready.

Required Qualifications

• BS, MS, or Ph.D. in Machine Learning, Statistics, Data Science, Robotics, Computer Science, or equivalent experience.
• Strong foundation in statistics, experimental design, machine learning evaluation, and data analysis.
• Strong Python and SQL skills and experience building reproducible analyses or evaluation systems.
• Ability to translate ambiguous behavior into measurable definitions without losing operational meaning.
• Excellent communication skills for explaining uncertainty, tradeoffs, and failure patterns to research and engineering teams.

Preferred Qualifications

• Experience evaluating robotic manipulation, autonomous systems, multimodal models, or safety-critical ML.
• Experience with causal inference, sequential experiments, uncertainty calibration, or human evaluation.
• Experience building annotation operations, data-quality programs, or active-learning systems.
• Experience with dashboards, experiment registries, model cards, or release governance.
• Experience linking offline metrics to field outcomes and identifying misleading proxies.

What We Offer

• Competitive compensation based on experience, level, and technical impact.
• The tools, compute, robots, and equipment needed to do your best work.
• The opportunity to build, test, and deploy directly on FRIDAY and the Holiday Robotics full stack.
• Close collaboration with researchers and engineers across hardware, control, simulation, learning, perception, data, and operations.

This is an evergreen talent-pool posting. We review profiles continuously, and the timing and level of a formal hiring process may depend on team priorities and available openings. If your work matches our direction, we would still like to hear from you.

Apply