Job Responsibilities

• 
  Develop and maintain offline and real-time data features to support the company’s risk control infrastructure.

• Build and manage comprehensive offline data models within the risk domain to empower analysts and algorithm teams in feature engineering and model development.

• Design and implement unified data pipelines that integrate offline and streaming architectures, enabling efficient real-time, mini-batch, and batch feature materialization for online indexes and feature services.

• Construct scalable frameworks combining offline and real-time data storage and processing to deliver reliable, high-quality feature data for risk analytics and decision-support platforms.

• Continuously enhance data processing efficiency, stability, and accuracy through systematic performance optimization and process automation.

  
  

Job Requirements

• 
  Proven experience with Apache Flink in large-scale big data environments, with successful delivery of end-to-end data projects.

• Solid understanding of HDFS architecture and operations; familiarity with Parquet or ORC columnar storage formats is an advantage.

• Hands-on experience with other big data frameworks such as Hive, HBase, or similar technologies is preferred.

• Strong background in data warehouse design and modeling, particularly in the risk management or middleware context.

• Proficiency in Flink SQL and Hive SQL, with extensive real-world implementation experience.

• Strong command of Java programming, with solid debugging and optimization skills.

• Ability to address complex and ambiguous technical problems with a proactive, learning-driven mindset; prior exposure to petabyte-scale data environments is a plus.

• Experience in building feature or sample platforms to support risk control analytics or model training workflows is a strong advantage.

• 5+ years of overall work experience, including at least 2 years in big data engineering or risk control-related fields.