Data Scientist

As a Data Scientist, you are the brain behind our threat-detection capabilities. Scammers evolve their tactics daily, moving from simple phishing links to highly sophisticated, multi-language social engineering attacks. Your mission is to stay one step ahead.

You will own the end-to-end data lifecycle, from building datasets that capture modern scam vectors to training and improving NLP/LLM-powered models that detect malicious, deceptive, and suspicious content with high precision. This role requires strong depth in both machine learning on structured/tabular data and natural language processing, as our detection systems rely on combining behavioral signals, metadata, and message content across multiple channels.

If you enjoy dissecting messy text data, working on multilingual detection problems, and building models that have an immediate, real-world protective impact, this is your role.

Key Responsibilities:

• Build and improve machine learning models for scam, fraud, and threat detection using both tabular and text-based data.
• Own the end-to-end data science workflow, including dataset construction, data preparation, feature development, model training, evaluation, and iteration.
• Develop and improve NLP/LLM-based models for detecting phishing, scam attempts, impersonation, and social engineering patterns across messages and communication content.
• Design multilingual detection approaches for scam and fraud scenarios across different languages and language-mixing patterns.
• Work with messy, evolving real-world data to identify new scam vectors and emerging attacker behaviors.
• Evaluate and optimize model performance with strong attention to precision, recall, class imbalance, robustness, and minimizing false positives.
• Partner closely with AI Engineers and product engineers to prepare models for real-world deployment.
• Conduct error analysis and continuous model improvement to stay ahead of changing scam tactics.
• Help define practical experimentation, labeling, and evaluation strategies for high-risk detection use cases.

Qualifications:

• Education: Bachelor’s degree/ Master’s degree or Ph.D. in Computer Science, Artificial Intelligence, Machine Learning, Electrical Engineering, or a related field.
• 5+ years of experience in Data Science or Applied Machine Learning.
• Strong hands-on experience with machine learning.
• Strong hands-on experience with NLP/LLM techniques, including text classification, transformer-based models, embeddings, fine-tuning, or related language understanding approaches.
• Experience working on multilingual language problems or text understanding across multiple languages.
• Strong Python skills and experience with common ML tooling and frameworks.
• Solid understanding of model evaluation, especially for imbalanced classification and high-precision detection problems.
• Experience building datasets and training pipelines from noisy, real-world data.
• Strong grounding in statistics, machine learning fundamentals, and practical model development.
• Ability to work closely with engineers to translate models into production-ready systems.

Nice to have:

• Experience with graph-based methods or graph-enhanced risk modeling.
• Experience in fraud detection, trust & safety, spam detection, abuse detection, or cybersecurity-related ML problems.
• Experience working with privacy-sensitive or risk-sensitive product domains.
• Familiarity with model monitoring, drift detection, and production ML workflows.