This role combines deep technical expertise with leadership responsibility, driving multidisciplinary Data Science initiatives end-to-end within complex security policy management systems.
The role drives the development of innovative, production-grade AI capabilities, including the intelligence behind security AI agents and advanced machine learning models built on complex security data.
Original thinking, deep technical rigor, intellectual agility, and exceptional problem-solving are essential.
Responsibilities:
- Lead end-to-end Data Science initiatives from problem framing through validation, CI/CD-based production deployment, monitoring, and ongoing operational optimization of AI systems
- Develop advanced ML capabilities, including predictive modeling, anomaly detection, classification, and behavioral analysis
- Develop the intelligent capabilities behind security AI agents, combining machine learning, LLMs, statistical methods, and domain-specific algorithms, with a strong understanding of how agents use these capabilities within multi-step workflows
- Adapt and fine-tune LLM technologies for domain-specific security use cases
- Define and implement rigorous evaluation methodologies for ML and agentic AI systems, including decision quality, reliability, robustness, uncertainty, and failure modes
- Partner with Product, Engineering, and Security teams to deliver measurable business impact
- Provide technical leadership and mentorship across multidisciplinary Data Science initiatives
Requirements
- M.Sc. in Data Science, Computer Science, Statistics, Mathematics, or a related quantitative discipline
- At least 7 years of hands-on Data Science experience, delivering end-to-end solutions into production environments
- Deep understanding of machine learning theory, statistical reasoning, and practical model behavior
- Strong expertise in Python and the modern Data Science ecosystem (NumPy, Pandas, Scikit-learn, PyTorch / TensorFlow, etc.)
- Strong understanding of LLM architectures, adaptation and fine-tuning methodologies
- Strong understanding of AI agent architectures and concepts, including tool use, context management, memory, planning/reasoning, and multi-step workflows
