Research Member of Technical Staff – Data & Evaluation

About the role

Rhoda AI is building generalist intelligent robots and seeks a Research Member of Technical Staff to lead data curation and evaluation for video action models. You will shape the quality of web‑scale video datasets and the metrics that drive model progress.

Key responsibilities

• Design and implement scalable pipelines for ingesting, deduplicating, filtering, and classifying web‑scale video data.
• Develop annotation frameworks and quality filters focused on motion quality, scene diversity, action content, and temporal coherence.
• Build evaluation benchmarks to measure video model capabilities such as prediction quality, temporal coherence, and downstream robot task performance.
• Research data selection, mixing, and weighting strategies to improve video generation quality and transfer to robotic control.
• Deploy vision‑language and video understanding models for automated annotation and content scoring at web scale.

Required profile

• Strong understanding of data‑centric machine learning and its impact on large generative models.
• Experience building large‑scale video data pipelines, including ingestion, filtering, deduplication, and quality scoring.
• Solid ML fundamentals with hands‑on experience training or evaluating large generative models.
• Ability to design reproducible, diagnostic evaluations for video generation models.

Required skills

• Data‑centric machine learning
• Large‑scale video data pipeline development
• Vision‑language models (VLMs) and video understanding models
• Evaluation framework design for video generation