Senior Data Scientist, OKX

• The right candidate will have strong intellectual curiosity and passion for achieving business results
• An ability to quickly define the problem, research and leverage state-of-the-art modeling techniques, and provide timely recommendations will be essential
• Key skills will include a strong analytical mindset, deep understanding of most popular machine learning algorithms and lead key initiatives with integrity and a passion for investigations, problem solving, and troubleshooting
• Master’s degree (or PhD) in Statistics, Mathematics, Operations Research, Computer Science, Economics, Engineering or other quantitative discipline. Bachelor’s degree with significant relevant experience will be considered
• 4+ years of fraud analytics experience in financial services or FinTechs
• Crypto/Blockchain experience
• Deep understanding of modern machine learning techniques / algorithms including GBM, XGBoost, LGBM, etc. Advanced programming skills of statistical / analytical software (SQL, R, Python,etc.)
• Successful track record of owning and driving large, complex data analysis projects
• Demonstrated capacity for innovation and outside-the-box thinking in the creation of new capabilities and processes that are unstructured or exploratory in nature
• Experience in a fast-paced startup environment with a strong level of initiative; and
• Ability and willingness to travel as needed
• Strong communicator in both writing and speaking
• Multi-tasking and strong project management skills

Desirable

• The Senior Data Scientist, Risk will offer a strategic perspective, deep analytical and modeling capabilities, and a collaborative working style
• Identify complex fraud patterns and their technical root causes through detailed data mining and analysis, including identification of sophisticated fraud methods employed by actors who are deliberately trying to avoid detection
• Serve as technical SME by sharing new data mining techniques, maintaining technical reference documentation, and interfacing with partner technology teams
• Collaborate across business and technology stakeholders to communicate analytical findings to both technical and non-technical audiences
• Provide technical guidance for engineering projects that incorporate new data points into the investigation team’s toolkit, such as API integrations or internal data transformations
• Link Analysis/Graph analytics to find and mitigate deeply-connected fraud networks and detect new accounts being added to these networks
• Unsupervised learning methods to augment existing supervised models, or detect portfolio anomalies
• Development of machine learning models
• Partner with product and engineering team in implementing features and models, and enhancing systems