TurbineOne

Data Scientist

• Ingesting, organizing, and maintaining large-scale training datasets from open-source resources and contract-specific artifacts
• Creating and managing data cataloging systems to ensure datasets are findable, accessible, and ready for ML training pipelines
• Designing and implementing data labeling workflows, including managing external labeling vendors and quality assurance processes
• Building and maintaining YOLO-style manifests and annotation formats for custom computer vision datasets
• Performing data cleaning, validation, and augmentation to ensure high-quality training data
• Conducting exploratory data analysis and generating insights about dataset characteristics, biases, and coverage gaps
• Supporting the ML research team with statistical analysis, experiment design, and model evaluation
• Developing data pipelines and automation tools for continuous data ingestion and processing
• Collaborating with ML engineers to optimize data loading and preprocessing for training efficiency
• Process incoming datasets from various sources, performing quality checks and organizing them into our data management system
• Create or review annotation schemas and coordinate with labeling teams to ensure consistent, high-quality labels
• Write Python scripts to clean, transform, and validate datasets for specific ML training requirements
• Analyze dataset statistics and create visualizations to identify potential issues or opportunities for improvement
• Collaborate with the ML research lead to design experiments and evaluate model performance across different data splits
• Document dataset characteristics, versioning, and lineage to maintain reproducibility and compliance

Senior Data Scientist, Machine Learning Data Operations