Key Responsibilities: Risk Control Model Development: Build user behavior models for forex trading scenarios, including classification models (to distinguish normal traders vs.
exploit/arbitrage abusers) and anomaly detection models.
Multi-dimensional Feature Engineering: Deeply mine massive transaction data and nonlinearly fuse real-time K-line trends, market volatility, and other environmental features with user trading behavior to construct a high-dimensional feature repository.
Order Routing Optimization: Use machine learning or reinforcement learning (RL) algorithms to optimize order routing decisions (A-Book external hedging vs.
B-Book internalization) for users with different risk levels, maximizing platform profitability.
Real-time Inference Deployment: Collaborate with engineering teams to deploy models in low-latency trading environments, ensuring millisecond-level order processing latency.
Strategy Iteration & Evaluation: Continuously monitor online model performance, analyze false positives/negatives, and enable model self-evolution and incremental training through feedback loops.
Requirements: Educational Background: Bachelor's degree or above in Computer Science, Mathematics, Statistics, Financial Engineering, or related fields; strong foundation in mathematics and algorithms.
Algorithm Expertise: Proficient in tree-based models such as Random Forest, XGBoost, and Light GBM; hands-on experience with time-series deep learning architectures like RNN/LSTM/Transformer.
Engineering Skills: Proficient in Python and scientific computing libraries (Pandas, Num Py, Scikit-learn); familiar with at least one deep learning framework (Py Torch/Tensor Flow); strong coding standards.
Data Processing: Familiar with real-time stream processing technologies such as Kafka, Flink, or Spark Streaming; capable of handling large-scale, high-frequency financial time-series data.
Industry Knowledge: Deep understanding of trading mechanisms for forex, gold, and other financial products; familiarity with common arbitrage strategies such as latency arbitrage and scalping is a plus.
Preferred Qualifications: Experience in high-performance / low-latency system development.
Prior experience in risk control or quantitative modeling at well-known brokerages, forex dealers, or high-frequency trading firms.
