Conduct research on cutting-edge sequence modeling and representation learning approaches, including Transformer-based architectures and transaction foundation models
Read, reproduce, and implement recent papers related to user behavior modeling, sequential recommendation, NLP-inspired pretraining, and large-scale representation learning
Build and optimize transaction sequence modeling pipelines using large-scale user behavioral and credit-related datasets
Explore improvements in model architecture, training objectives, and feature representations to enhance downstream risk prediction performance
Collaborate with risk modeling, data engineering, and strategy teams to evaluate and productionize research findings into real-world risk applications
Requirements
Undergraduates from a degree in Computer Science, Business Analytics and related fields
Full-time interns preferred (3 to 6 months)
Past internship experience in sequence modelling or NLP related model training is compulsory
Strong programming skills in Python; familiarity with SQL and Spark/Hadoop is a plus
Good understanding of machine learning and deep learning fundamentals, especially sequence modeling and Transformer architectures
Hands-on experience with deep learning frameworks such as PyTorch or TensorFlow
Strong interest in large-scale AI models, representation learning, NLP-inspired modeling, or recommendation systems
Ability to read and understand research papers and reproduce experimental results independently
Strong analytical thinking and curiosity toward applying advanced AI techniques to real-world risk problems
