Data Scientist

We are inviting a passionate Data Scientist to join our project — an AI-powered system for online fraud and scam detection. This innovative solution is designed to identify and prevent malicious activity across social media, including scams tied to investment schemes. Our goal is to uncover scam networks, analyze their behavior and money flows, and provide actionable insights to help block the financial channels used for moving illicit funds. Collaborating with AI engineers, developers, and security experts, you will tackle complex challenges by building models that reduce online scams and help make the internet a safer place for everyone.

Key Responsibilities:

• Design, develop, and evaluate machine learning models to detect and analyze malicious patterns in different types of data (textual, numerical, relational, etc.)
• Conduct thorough error analysis and optimize models for accuracy and efficiency
• Collaborate with cross-functional teams to improve data collection methods and enhance the datasets quality and diversity

Technical Stack:

• NLP: from-scratch models, fine-tuning, and LLM integration (agents/chatbots)
• Classical ML: tabular data with a timeline aspect
• Ruled-based systems: scoring systems tuned with ML methods
• MLOps: model monitoring, automated retraining pipelines (in collaboration with other teams)

 Main requirements:

• 3+ years of hands-on experience in AI development
• Strong Analytical Skills in Data Science: Proficiency in exploratory data analysis (EDA), feature engineering, and feature selection. The candidate must be capable of handling complex, non-human-readable data with hidden patterns, creating custom solutions, and avoiding errors that could significantly impact results
• Knowledge of Classical Data Science Methods: Understanding classical models, including their advantages and disadvantages, and the ability to propose justified solutions for their application
• Familiarity with NLP Techniques and Deep Learning: Familiarity with fundamental Natural Language Processing (NLP) techniques, experience with deep learning libraries, understanding the model training process, and having practical experience in training custom models
• Proficiency in English (Upper-Intermediate or higher), both written and spoken

Will be a plus:

• A degree in Computer Science, Math, or other related fields
• Basic understanding of Web3 technology and its components
• Experience with Graph Neural Network