Note: The job is a remote job and is open to candidates in USA. Varonis is seeking a Senior Data Scientist to develop and optimize machine learning and large language model capabilities for complex cybersecurity challenges. The role involves designing, evaluating, and deploying ML and LLM-based solutions, collaborating with cybersecurity researchers and engineering teams, and addressing complex data science problems from research through production.
Responsibilities
- Analyze large, complex datasets to identify patterns and trends that could indicate potential security threats
- Collaborate with cybersecurity researchers to understand emerging threats and develop solutions that leverage advanced machine learning and data analytics
- Design, develop, and evaluate ML and LLM-based solutions, including classification, NLP, and security-focused models, from experimentation through production
- Develop and optimize LLM techniques including prompting, retrieval, embeddings, RAG, and fine-tuning where appropriate
- Design robust evaluation methodologies and experiments for classification and LLM-based systems, using appropriate metrics to measure model quality, accuracy, and performance
- Stay up to date with advancements in machine learning, NLP, LLMs, and related technologies, and evaluate their applicability to real-world cybersecurity problems
- Collaborate with software and data engineers to integrate ML and LLM solutions into scalable production systems
- Take ownership of complex data science problems, from problem definition and research through implementation and evaluation
Skills
- Bachelor's degree in Computer Science, Engineering, Statistics, Mathematics, or a related field
- 5+ years of hands-on experience in data science, machine learning, or a related field
- Strong understanding of machine learning and deep learning, with hands-on experience developing and evaluating classification models
- Hands-on experience with NLP and text classification, including NER, training and fine-tuning models, and working with both traditional ML and modern LLM-based approaches
- Strong proficiency in Python
- Experience developing and evaluating LLM-based applications, including prompt engineering
- Experience with embeddings, vector search, and vector databases
- Experience working with commercial and open-source LLMs and frameworks, such as OpenAI models and the Hugging Face ecosystem
- Experience with model fine-tuning, RAG systems, and LLM-based classification
- Experience working with highly imbalanced datasets, selecting appropriate evaluation metrics, and optimizing models for precision/recall trade-offs
- Strong analytical and problem-solving skills, with the ability to evaluate and interpret complex data and experimental results
- Excellent communication and collaboration skills, with the ability to work effectively in a multidisciplinary team
- Proven track record of delivering high-quality solutions in a fast-paced, dynamic environment
- A Master's or Ph.D. is a plus
- Experience with PySpark and Databricks is a plus
Benefits
- Hybrid work arrangement
Company Overview
Company H1B Sponsorship