Senior Product Analyst

Department: Analytics Department

Employment Type: Full Time

Location: Kazakhstan

Description

Key Responsibilities

• Own antifraud solutions end-to-end, from problem framing and feature design to production deployment, monitoring, and iteration based on real-world performance.
• Design, build, and productionize machine learning and rule-based antifraud models, integrating them into existing real-time and batch decision flows.
• Take ownership of antifraud models after release, including rollout strategy, performance monitoring, degradation detection, and rollback when required.
• Continuously analyze fraud patterns, false positives, and false negatives to improve detection quality and minimize negative impact on legitimate users.
• Develop and maintain reliable feature pipelines for antifraud use cases, working with large-scale event and user-level data.
• Ensure antifraud solutions are production-ready, handling data quality issues, missing or delayed signals, evolving fraud behavior, and concept drift.
• Collaborate closely with Product Analysts, backend engineers, and Trust & Safety stakeholders to align ML decisions with business and operational needs.

Skills, Knowledge and Expertise

• 4+ years of experience in applied machine learning, data science, or ML engineering, with a strong focus on fraud detection, abuse prevention, or behavioral analysis.
• Proven experience owning antifraud solutions end-to-end, from problem definition and modeling to production deployment and iteration beyond offline experiments.
• Strong Python engineering skills, with hands-on experience writing production-oriented code for data processing, feature engineering, and model inference.
• Practical experience with Python DS and ML libraries including pandas, NumPy, scikit-learn, and gradient boosting frameworks (e.g., LightGBM or XGBoost).
• Experience working with highly imbalanced datasets and understanding fraud-specific trade-offs between precision, recall, and business impact.
• Hands-on experience with model evaluation and interpretability, including tools such as SHAP or equivalent approaches.
• Experience analyzing large-scale event or user-level data in batch or near–real-time environments.
• Ability to independently structure complex, ambiguous antifraud problems and deliver practical, scalable, production-ready solutions.
• Strong collaboration and communication skills, with experience working effectively alongside Product Analysts, engineering teams, and other Data Science practitioners.

Conditions

• Stable salary, official employment
• Health insurance
• Hybrid work mode and flexible schedule
• Access to professional counseling services, including psychological, financial, and legal support
• Discount club membership
• Diverse internal training programs
• Partially or fully paid additional training courses
• All necessary work equipment