Algorithm Engineer, Marketplace Intelligence & Data - Computer Vision (2026 Graduate)

• Develop CV-based identity verification models and services, including:
  • Anti-spoofing against various attack types (e.g., deepfake)
  • Passive and active liveness checks
  • OCR for IC cards and identity documents
  • Anti-spoofing for IC card tampering
  • On-device and server-side image quality assessment.
• Design and implement robust training, inference, and data processing pipelines.
• Write high quality, elegant, readable, and easy to maintain code.
• Collaborate with business stakeholders, product managers, and engineering team, and get involved in the entire life cycle of our eKYC solutions, from problem formulation, technical design, to implementation, upgrading, and maintenance.

• Bachelor or above in Computer Science, Machine Learning, or relevant majors.
• Proficient in Python and deep learning framework (e.g., PyTorch) to build custom modules, dataloaders, and training loops.
• Deep and broad knowledge in at least one of areas: computer vision, multimodel, or vision-language models (VLMs).
• Familiar with model debugging, dynamic computation graphs, and memory management.
• Familiar with large-scale model development frameworks (e.g., HuggingFace, LLaMA-Factory) and deployment pipelines (e.g., ONNX, TorchScript).
• Familiar with model optimization techniques (e.g., pruning, quantization, knowledge distillation) and training efficiency techniques (e.g., mixed precision, distributed training).
• Good teamworking spirit, positive attitude, self-driven.
• Good oral and written English, good communication skills.
• Passionate about technology, fast learner, willing to share.

Good To Have:

• Master's or above in Computer Science or relevant Majors.
• Publications in top-tier computer vision or AI conferences and journals.
• Prior experience in eKYC, digital identity, fraud prevention or security-related projects.
• Prior experience in VLMs/LLMs.