Senior AI Data Engineer, SentinelOne

• 5+ years of experience working with data in applied AI or ML systems, ideally in security or similarly complex domains
• Strong Python skills and experience with common data tools (pandas, numpy, PySpark)
• Experience curating and maintaining large, evolving datasets for training and evaluation
• Familiarity with data labeling strategies, data versioning, and pipeline automation
• Ability to explore and analyze messy data from real-world systems
• Experience with LLMs or generative AI models, and understanding of their data dependencies
• A working knowledge of cybersecurity concepts, threat types, telemetry sources, or detection workflows

Desirable

• We’re looking for a data scientist who’s deeply motivated to build high-quality datasets that power cybersecurity-focused AI
• Your primary mission will be to design and maintain the data foundations behind products that detect threats, summarize alerts, and enable defenders
• You should be energized by ambiguity, eager to explore noisy and complex security data, and experienced in transforming it into structured, useful assets
• While your main focus will be on data—collecting, cleaning, labeling, and maintaining—you’ll also help define how we measure quality and track progress over time
• We’re especially excited about candidates who understand the nuances of cybersecurity and want to shape how AI is applied to real-world threats
• Build and maintain datasets that power training, fine-tuning, and prompting of AI models for cybersecurity applications
• Explore and curate structured and unstructured security data (logs, malware, alerts, incidents) to improve model relevance and effectiveness
• Develop pipelines for data collection, cleaning, labeling, and versioning
• Collaborate with researchers and engineers to define data requirements and ensure relevance to cybersecurity problems
• Contribute to evaluation design and automation to ensure continuous improvement of AI systems
• Drive data quality and coverage for use cases like summarization, classification, detection, and reasoning over security context